Band 33, Heft 1

In order to enhance the performance of machine translation, this article briefly introduced algorithms that can be used to extract semantic feature vectors. Then, the aforementioned algorithms were integrated with the encoder–decoder translation algorithm, and the resulting algorithms were subsequently tested. First, the performance of the semantic recognition of the long short-term memory (LSTM)-based semantic feature extractor was tested, followed by a comparison with the translation algorithm that does not include semantic features, as well as the translation algorithm that incorporates convolutional neural network-extracted semantic features. The findings demonstrated that the LSTM-based semantic feature extractor accurately identified the semantics of the source language. The proposed translation algorithm, which is based on LSTM semantic features, achieved more accurate translations compared to the other two algorithms. Furthermore, it was less affected by the length of the source language.

The present study examined the usage of deep convolutional neural networks (DCNNs) for the classification, segmentation, and detection of the images of surface defects in heritage buildings. A survey was conducted on the building surface defects in Gulang Island (a UNESCO World Cultural Heritage Site), which were subsequently classified into six categories according to relevant standards. A Swin Transformer- and YOLOv5-based model was built for the automated detection of surface defects. Experimental results suggested that the proposed model was 99.2% accurate at classifying plant penetration and achieved a mean intersection-over-union (mIoU) of over 92% in relation to moss, cracking, alkalization, staining, and deterioration, outperforming CNN-based semantic segmentation networks such as FCN, PSPNet, and DeepLabv3plus. The Swin Transformer-based approach for the segmentation of building surface defect images achieved the highest accuracy regardless of the evaluation metric (with an mIoU of 90.96% and an mAcc of 95.78%), when contrasted to mainstream DCNNs such as SegFormer, PSPNet, and DANet.

Although traditional image classification techniques are often used in authentic ways, they have several drawbacks, such as unsatisfactory results, poor classification accuracy, and a lack of flexibility. In this study, we introduce a combination of convolutional neural network (CNN) and support vector machine (SVM), along with a modified bag of visual words (BoVW)-based image classification model. BoVW uses scale-invariant feature transform (SIFT) and Oriented Fast and Rotated BRIEF (ORB) descriptors; as a consequence, the SIFT–ORB–BoVW model developed contains highly discriminating features, which enhance the performance of the classifier. To identify appropriate images and overcome challenges, we have also explored the possibility of utilizing a fuzzy Bag of Visual Words (BoVW) approach. This study also discusses using CNNs/SVM to improve the proposed feature extractor’s ability to learn more relevant visual vocabulary from the image. The proposed technique was compared with classic BoVW. The experimental results proved the significant enhancement of the proposed technique in terms of performance and accuracy over state-of-the-art models of BoVW.

Internet has become the primary source of extracurricular entertainment for college students in today’s information age of Internet entertainment. However, excessive Internet addiction (IA) can negatively impact a student’s daily life and academic performance. This study used Stochastic models to gather data on campus education behaviour, extract the temporal characteristics of university students’ behaviour, and build a Stochastic dropout long short-term memory (LSTM) network by fusing Dropout and LSTM algorithms in order to identify and analyse the degree of IA among university students. The model is then used to locate and forecast the multidimensional vectors gathered, and finally to locate and evaluate the extent of university students’ Internet addiction. According to the experiment’s findings, there were 4.23% Internet-dependent students among the overall (5,861 university students), and 95.66% of those students were male. The study examined the model using four dimensions, and the experimental findings revealed that the predictive model suggested in the study had much superior predictive performance than other models, scoring 0.73, 0.72, 0.74, and 0.74 on each dimension, respectively. The prediction model outperformed other algorithms overall and in the evaluation of the four dimensions, performing more evenly than other algorithms in the performance comparison test with other similar models. This demonstrated the superiority of the research model.

The quality of the output produced by the multi-layer perceptron neural network depends on the careful selection of its weights and biases. The gradient descent technique is commonly used for choosing MLP’s optimal configuration, but it can suffer from being stuck in local optima and slow convergence toward promising regions in the search space. In this article, we propose two new optimization algorithms based on the moth-flame optimization algorithm (MFO), which mimics moths’ special navigation methods at night. We use these algorithms to enhance the performance of the training process of the MLP neural network. To demonstrate the effectiveness of our approach, we apply it to the problem of predicting iron ore prices, which plays an important role in the continuous development of the steel industry. We use a large number of features to predict the iron ore price, and we select a promising set of features using two feature reduction methods: Pearson’s correlation and a newly proposed categorized correlation. Surprisingly, new features not mentioned in the literature are discovered, and some are discarded. The time series dataset used has been extracted from several sources and pre-processed to fit the proposed model. We compare our two proposed MFO algorithms, the roulette wheel moth-flame optimization algorithm and the global best moth-flame optimization algorithm, against four swarm intelligence algorithms and five classical machine learning techniques when predicting the iron ore price. The results acquired indicate the superior performance of the suggested algorithms concerning prediction accuracy, root-mean-square error, mean-square error, average absolute relative deviation, and mean absolute error. Overall, our work presents a promising approach for improving the performance of MLP neural networks, and it demonstrates its effectiveness in the challenging problem of predicting iron ore prices.

Bitcoin (BTC) is one of the most important cryptocurrencies widely used in various financial and commercial transactions due to the fluctuations in the price of this currency. Recent research in large data analytics and natural language processing has resulted in the development of automated techniques for assessing the sentiment in online communities, which has emerged as a crucial platform for users to express their thoughts and comments. Twitter, one of the most well-known social media platforms, provides many tweets about the BTC cryptocurrency. With this knowledge, we can apply deep learning (DL) to use these data to predict BTC price variations. The researchers are interested in studying and analyzing the reasons contributing to the BTC price’s erratic movement by analyzing Twitter sentiment. The main problem in this article is that no standard model with high accuracy can be relied upon in analyzing textual emotions, as it represents one of the factors affecting the rise and fall in the price of cryptocurrencies. This article aims to classify the sentiments of an expression into positive, negative, or neutral emotions. The methods that have been used are word embedding FastText model in addition to different DL methods that deal with time series, one-dimensional convolutional neural networks (CONV1D), long-short-term memory networks (LSTMs), recurrent neural networks, gated recurrent units, and a Bi-LSTM + CONV1D The main results revealed that the LSTM method, based on the DL technique, achieved the best results. The performance accuracy of the methods was 95.01, 95.95, 80.59, 95.82, and 95.67%, respectively. Thus, we conclude that the LSTM method achieved better results than other methods in analyzing the textual sentiment of BTC.

To prevent in a short time the rapid increase of tourists and corresponding traffic restriction measures’ lack in scenic areas, this study established a prediction model based on an improved convolutional neural network (CNN) and long- and short-term memory (LSTM) combined neural network. The study used this to predict the inflow and outflow of tourists in scenic areas. The model uses a residual unit, batch normalization, and principal component analysis to improve the CNN. The experimental results show that the model works best when batches’ quantity is 10, neurons’ quantity in the LSTM layer is 50, and the number of iterations is 50 on a workday; on non-working days, it is best to choose 10, 100, or 50. Using root mean square error (RMSE), normalized root mean square error (NRMSE), mean absolute percentage error (MAPE), and mean absolute error (MAE) as evaluation indicators, the inflow and outflow RMSEs of this study model are 82.51 and 89.80, MAEs are 26.92 and 30.91, NRMSEs are 3.99 and 3.94, and MAPEs are 1.55 and 1.53. Among the various models, this research model possesses the best prediction function. This provides a more accurate prediction method for the prediction of visitors’ flow rate in scenic spots. Meanwhile, the research model is also conducive to making corresponding flow-limiting measures to protect the ecology of the scenic area.

Purpose The purpose of this research article is to analyze the short-term wind speed and develop a framework model to overcome the challenges in the wind power industry. Design/Methodology/Approach Real data with a case study of wind speed is presented to illustrate the advantages of this new wind speed analytical framework. Hourly measurements of wind speed are observed, and the experiments are conducted using tools such as ANOVA, control charts, trend analysis, and predictive models. The August month data for over 13 years from modern era retrospective-analysis for research and applications (MERRA) National aeronautics and space administration (NASA) for Coimbatore and Erode locations in Tamil Nadu, India, have been used. The results were considered for the study to understand the wind speed data and the implementation of new wind power projects in India. Findings The essence of the proposed wind speed analytical framework is its flexible approach, which enables the effective integration of wind firms’ individual requirements by developing tailor-made analytical evaluations. Originality/Value This article derives the wind speed analytical framework with the application of statistical tools and machine learning algorithms.

Accurate early identification and treatment of cardiovascular diseases can prevent heart failure problems and reduce mortality rates. This study aims to use quantum learning to predict heart problems to increase the accuracy of traditional prediction and classification methods. Machine learning (ML) and deep learning (DL) techniques need quantum learning to quickly and accurately analyze massive volumes of complex data. With quantum computing, the suggested DL and ML algorithms can change their predictions on the basis of changes in the dataset. This approach could help with the early and accurate detection of chronic diseases. The Cleveland heart disease dataset is undergoing preliminary processing to validate missing values to increase the precision rate and prevent incorrect forecasts. This study examined the feasibility of employing and deploying a quantum ML (QML) framework via cloud computing to categorize cardiac conditions. The research was divided into four sections. First, the principal component analysis was used to preprocess the Cleveland dataset, recursive feature elimination was used to select features, and min–max normalization was used to give the dataset a high-dimensional value. Second, we compared traditional classifiers, such as support vector machine (SVM) and artificial neural network, with the quantum approach to verify the quantum approach’s efficiency. Third, we examined two unique QML classification methods: quantum neural networks (QNNs) and quantum SVM (QSVM). Fourth, bagging-QSVM was developed and deployed as an ensemble learning model. Experimental results using the QNN show an accuracy of 77%, a precision of 76%, a recall of 73%, and an F 1 score of 75%. With an accuracy of 85%, a precision of 79%, a recall of 90%, and an F 1-score of 84%, the QSVM method demonstrated a much better performance than the QNN. Particularly, the Bagging_QSVM model exhibited an outstanding performance, with a flawless score of 100% across all critical performance measures. The study shows that the bagging method for ensemble learning is a solid way of increasing the accuracy of quantum method predictions.

Teaching quality evaluation is a process of evaluating the teaching quality of architectural majors. It can not only evaluate the teaching level of teachers, but also evaluate the learning effectiveness of students. Therefore, this study designs a teaching quality evaluation system for architecture majors based on fuzzy environment, in order to provide direction guidance for effectively evaluating the teaching quality of architecture majors by using this research. The teaching quality evaluation of higher vocational architecture majors based on enterprise platform is a multiple-attribute group decision-making (MAGDM). The spherical fuzzy sets (SFSs) provide more free space for decision makers to portray uncertain information during the teaching quality evaluation of higher vocational architecture majors based on enterprise platform. Therefore, this study expands the partitioned Maclaurin symmetric mean operator and induced ordered weighted average operator to SFSs based on the power average technique and construct induced spherical fuzzy power partitioned MSM (I-SFPPMSM) technique. Subsequently, a novel MAGDM method is put forward based on I-SFPPMSM technique and spherical fuzzy number weighted geometric technique under SFSs. Finally, a numerical example for teaching quality evaluation of higher vocational architecture majors based on enterprise platform is employed to verify the put forward method, and comparative analysis with some existing techniques to testy the validity and superiority of the I-SFPPMSM technique.

Sickle cell disease (SCD), a blood disorder that transforms the shape of red blood cells into a distinctive sickle form, is a major concern as it not only compromises the blood’s oxygen-carrying capacity but also poses significant health risks, ranging from weakness to paralysis and, in severe cases, even fatality. This condition not only underscores the pressing need for innovative solutions but also encapsulates the broader challenges faced by medical professionals, including delayed treatment, protracted processes, and the potential for subjective errors in diagnosis and classification. Consequently, the application of artificial intelligence (AI) in healthcare has emerged as a transformative force, inspiring multidisciplinary efforts to overcome the complexities associated with SCD and enhance diagnostic accuracy and treatment outcomes. The use of transfer learning helps to extract features from the input dataset and give an accurate prediction. We analyse and compare the performance parameters of three distinct models for this purpose: GoogLeNet, ResNet18, and ResNet50. The best results were shown by the ResNet50 model, with an accuracy of 94.90%. Explainable AI is the best approach for transparency and confirmation of the predictions made by the classifiers. This research utilizes Grad-CAM to interpret and make the models more reliable. Therefore, this specific approach benefits pathologists through its speed, precision, and accuracy of classification of sickle cells.

As an effective tool to show the fuzziness of qualitative information, the interval-valued T-spherical fuzzy set can utilize three kinds of information, namely, membership, abstinence, and non-membership, to show the opinions of decision-maker. Given this advantage, many interval-valued T-spherical fuzzy multi-criteria decision-making (IVTSF-MCDM) methods have been designed. However, most of the existing IVTSF-MCDM methods have a common limitation that the inability to effectively show the impacts of extreme data. To address this limitation, this study develops a novel MCDM method based on interval-valued T-spherical fuzzy extended power aggregation operator. First, interval-valued T-spherical fuzzy cross-entropy (CE) and interval-valued T-spherical fuzzy symmetrical CE are defined to measure the difference between two interval-valued T-spherical fuzzy numbers, which are used to determine criteria weights in MCDM. Second, interval-valued T-spherical fuzzy extended power average operator and interval-valued T-spherical fuzzy extended power geometric operator are proposed, and their properties are investigated. Moreover, in view of that criteria may be assigned to different weights, this study defines interval-valued T-spherical fuzzy extended power weighted average operator and interval-valued T-spherical fuzzy extended power weighted geometric operator to derive the order of alternatives. Finally, the applicability of the proposed method is validated by the case about investment country selection, while the sensitivity and comparison analyses are also conducted to further prove its advantages and effectiveness.

Neighborhood relationships play a pivotal role in rough set theory, addressing the limitations of equivalence relations. This article focuses on defining upper and lower approximation operators using neighborhood relationships and exploring their properties in terms of serialization, inverse serialization, reflexivity, symmetry, transitivity, and Euclidean relations. Furthermore, a necessary and sufficient condition for the upper approximation operator to function as a topological closure operator is derived. Overall, this research sheds light on the significance of neighborhood relationships and their implications within rough set theory.

With the refinement and scientificization of sports training, the demand for sports performance analysis in the field of sports has gradually become prominent. In response to the problem of low accuracy and poor real-time performance in human pose estimation during sports, this article focused on volleyball sports and used a combination model of OpenPose and DeepSORT to perform real-time pose estimation and tracking on volleyball videos. First, the OpenPose algorithm was adopted to estimate the posture of the human body region, accurately estimating the coordinates of key points, and assisting the model in understanding the posture. Then, the DeepSORT model target tracking algorithm was utilized to track the detected human pose information in real-time, ensuring consistency of identification and continuity of position between different frames. Finally, using unmanned aerial vehicles as carriers, the YOLOv4 object detection model was used to perform real-time human pose detection on standardized images. The experimental results on the Volleyball Activity Dataset showed that the OpenPose model had a pose estimation accuracy of 98.23%, which was 6.17% higher than the PoseNet model. The overall processing speed reached 16.7 frames/s. It has good pose recognition accuracy and real-time performance and can adapt to various volleyball match scenes.

To address the challenges of secure and reliable communication and system quality of service (QoS) prediction in intelligent production lines (IPL) in the Industrial Intelligent Internet of Things (II-IOT) environment, a redundant collaborative security model-based communication architecture is designed. First, the redundant collaborative security communication model is introduced to construct the network communication architecture of IPL, including the industrial-site mechanical floor, data awareness layer, and gateway and application layer. Then, to leverage the advantages of the empirical-mode decomposition (EMD) method and the bidirectional long short-term memory (BiLSTM) model in time-series data analysis and processing, an EMD-BiLSTM-based QoS prediction model is proposed that can synchronously achieve one-step and multi-step prediction of QoS attributes. The proposed model exhibits a prediction accuracy of up to 94.01% on the SourceForge dataset, with prediction, recall, and F 1 values as high as 91.37, 90.60, and 90.99%, respectively. The proposed EMD-BiLSTM model can achieve better performance than the state-of-the-art QoS prediction models, indicating that the proposed model can be more effectively used to improve the reliable communication level of II-IoT.

In the context of “Internet plus,” college English-teaching resources are increasingly rich. Research has found that implementing a blended teaching model for college English based on the Super Star Learning Communication and Rain Classroom online teaching platform is beneficial for improving students’ enthusiasm and continuity in English learning. The English-blended teaching quality evaluation is a multiple attribute decision making (MADM). The single-valued neutrosophic set (SVNS) is a useful tool to depict uncertain information during the English-blended teaching quality evaluation. In such an article, the single-valued neutrosophic number Aczel–Alsina power geometric (SVNNAAPG) operator is produced based on the Aczel–Alsina operations and classical power geometric operator under SVNSs. The SVNNAAPG operator is built for MADM. Eventually, an example about English-blended teaching quality evaluation and some selected comparative analysis was used to depict the SVNNAAPG technique.

This article proposes an English grammar intelligent error correction model based on the attention mechanism and Recurrent Neural Network (RNN) algorithm. It aims to improve the accuracy and effectiveness of error correction by combining the powerful context-capturing ability of the attention mechanism with the sequential modeling ability of RNN. First, based on the improvement of recurrent neural networks, a bidirectional gated recurrent network is added to form a dual encoder structure. The encoder is responsible for reading and understanding the input text, while the decoder is responsible for generating the corrected text. Second, the attention mechanism is introduced into the decoder to convert the output of the encoder into the attention probability distribution for integration. This allows the model to focus on the relevant input word as it generates each corrected word. The results of the study showed that the model was 2.35% points higher than statistical machine translation–neural machine translation in the CoNLL-2014 test set, and only 1.24 points lower than the human assessment score, almost close to the human assessment level. The model proposed in this study not only created a new way of English grammar error correction based on the attention mechanism and RNN algorithm in theory but also effectively improved the accuracy and efficiency of English grammar error correction in practice. It further provides English learners with higher-quality intelligent error correction tools, which can help them learn and improve their English level more effectively.

In response to the existing problems in the prediction accuracy, data collection, real-time monitoring, and consideration of factors leading to depression in the current mechanism for predicting depression among college students, this article used computer intelligent systems to study the prediction mechanism of depression among college students. This article conducted a survey on students at University A using a survey questionnaire to understand the main reasons that affect their tendency to develop depression. It processed and analyzed the data using the Beck Depression Scale and Statistical Product and Service Solution 21.0 (SPSS 21.0). Meanwhile, natural language processing techniques in computer intelligence systems can be utilized. This article combines emotional dictionaries and word frequency-inverse document frequency to construct a prediction mechanism model for depression tendencies among college students, improving the accuracy of predicting student depression tendencies. The experiment shows that the average accuracy of the depression tendency prediction mechanism model constructed based on Natural Language Processing technology after 50 experiments was 97.02%, which was 5.33% higher than the model constructed based on neural network calculations. Overall, research on the prediction mechanism of depression tendency among college students based on computer intelligence systems can provide more effective mental health support and intervention measures for schools, helping students improve their psychological state, academic achievement, and quality of life.

This study provides a concise overview of a grammatical error correction algorithm that is based on an encoder-decoder machine translation structure. Additionally, it incorporates the attention mechanism to enhance the algorithm’s performance. Subsequently, simulation experiments were conducted to compare the improved algorithm with an algorithm based on a classification model and an algorithm based on the traditional translation model using open corpus data and English translations from freshmen. The results demonstrated that the optimized algorithm yielded superior intuitive error correction outcomes. When applied to both the open corpus and the English translations of college freshmen, the optimized error correction algorithm outperformed the others. The traditional translation model-based algorithm came in second, while the classification model-based algorithm showed the least favorable performance. Furthermore, all three error correction algorithms experienced a decrease in performance when dealing with English compositions from freshmen. However, the optimized algorithm exhibited a relatively smaller decline.

Microblogs are currently one of the most well-liked social platforms in China, and sentiment analysis of microblog texts can help further analyze the realization of their media value; however, the current task of sentiment analysis based on microblog information suffers from low accuracy due to the large size and high redundancy of microblog data, a microblog sentiment analysis method using Bidirectional Encoder Representation from Transformers (BERT)–Text Convolutional Neural Network (TextCNN)–Bidirectional Gate Recurrent Unit (BiGRU)–Multihead-Attention model in Spark big data environment is proposed. First, the Chinese pre-trained language model BERT is used to convert the input data into dynamic character-level word vectors; then, TextCNN is used to effectively obtain local features such as keywords and pool the filtered features; then, BiGRU is introduced to quickly capture more comprehensive semantic information; finally, a multi-headed attention mechanism is implemented to emphasize the most significant features in order to accomplish the sentiment classification of microblog information task precisely. By comparing the existing advanced models, the proposed model demonstrates an improvement of at least 4.99% and 0.05 in accuracy and F1-score evaluation indexes, respectively. This enhancement significantly enhances the accuracy of microblog sentiment analysis tasks and aids pertinent authorities in comprehending the inclination of individual’s attitude toward hot topics. Furthermore, it facilitates a prompt prediction of topic trends, enabling them to guide public opinion accordingly.

Nowadays, major enterprises and schools vigorously promote the combination of information technology and subject teaching, among which automatic grading technology is more widely used. In order to improve the efficiency of English composition correction, the study proposes an unsupervised semantic space model for English composition tangent, using a Hierarchical Topic Tree Hybrid Semantic Space to achieve topic representation and clustering in English composition; adopts a feature dimensionality reduction method to select a set of semantic features to complete the optimization of the feature semantic space; and combines the tangent analysis algorithm to achieve intelligent scoring of English composition. The experimental data show that the accuracy and F -value of the English composition tangent analysis method based on the semantic space are significantly improved, and the Pearson correlation coefficient between the unsupervised semantic space English composition tangent model and the teacher’s manual grading is 0.8936. The results show that the unsupervised semantic space English composition tangent model has a higher accuracy rate, is more applicable, and can efficiently complete the English composition grading task: essay review task.

Cardiotocography (CTG) is considered the gold standard for monitoring fetal heart rate (FHR) during pregnancy and labor to estimate the danger of oxygen deprivation. Visual interpretation of CTG traces is complex and frequently results in high rates of false positives and false negatives, leading to unfavorable and unwanted outcomes such as fetal mortality or needless cesarean surgery. If the data are well-balanced, which is uncommon in medical datasets, machine learning techniques can be helpful in interpretation. This study is designed to determine classification performance under various data balance approaches. We propose a robust methodology for the automated extraction of features that use a deep learning model based on the one-dimensional convolutional neural network (1D-CNN). We used a public database containing 552 intrapartum CTG recordings. Due to the imbalance in the dataset, the experiments were conducted under a variety of conditions such as (i) an unbalanced dataset, (ii) undersampling, (iii) a weighted binary cross-entropy approach, and (iv) oversampling utilizing the synthetic minority oversampling technique (SMOTE). We found an excellent sensitivity (99.80% for the unbalanced dataset, 96.25% for the weighted binary cross-entropy approach, and 99.81% with SMOTE) except for the under sampling situation, in which the sensitivity was 85.71%. Moreover, the 1D-CNN model incorporating SMOTE yielded promising results in 88% specificity, 93.72% quality index (QI), and 95.10% area under the curve. The model exhibited excellent performance in terms of sensitivity in every scenario except for undersampling. The oversampling of training data with SMOTE yielded a decent level of specificity, demonstrating the model’s strong predictive capacity. In addition, the SMOTE scenario resulted in fewer training epochs, which is another accomplishment.

This work addresses the challenges of scene segmentation and low segmentation accuracy in short videos by employing virtual reality (VR) technology alongside a 3D DenseNet model for real-time segmentation in dynamic scenes. First, this work extracted short videos by frame and removed redundant background information. Then, the volume rendering algorithm in VR technology was used to reconstruct short videos in dynamic scenes in 3D. It enriched the detailed information of short videos, and finally used the 3D DenseNet model for real-time segmentation of short videos in dynamic scenes, improving the accuracy of segmentation. The experiment compared the performance of High resolution network, Mask region based convolutional neural network, 3D U-Net, Efficient neural network models on the Densely annotation video segmentation dataset. The experimental results showed that the segmentation accuracy of the 3D DenseNet model has reached 99.03%, which was 15.11% higher than that of the ENet model. The precision rate reached 98.33%, and the average segmentation time reached 0.64 s, improving the segmentation accuracy and precision rate. It can adapt to various scene situations and has strong robustness. The significance of this research lies in its innovative approach in tackling these issues. By integrating VR technology with advanced deep learning models, we can achieve more precise segmentation of dynamic scenes in short videos, enabling real-time processing. This has significant practical implications for fields such as video editing, VR applications, and intelligent surveillance. Furthermore, the outcomes of this research contribute to advancing computer vision in video processing, providing valuable insights for the development of future intelligent video processing systems.

In traditional psychological counseling, limitations of patients’ subjective feelings and expressive abilities, as well as the varying levels of professional skills and conversation skills of counselors, make it difficult to properly solve psychological problems. Emotional recognition technology can help psychological counselors better understand the inner world of patients, thereby providing more targeted psychological counseling services for college students. 2D convolutional neural network-long short-term memory (2DCNN-LSTM) and 1D convolutional neural network-long short-term memory network (1DCNN-LSTM) were used for emotion recognition. Wavelet packet coefficient sequences were utilized to analyze multimodal data; multi-core convolutional neural network models were applied for multimodal feature layer fusion, and LSTM recurrent neural networks were used to effectively identify valence and college student emotions. The experimental results showed that under multimodal recognition, the average recognition accuracy of 2DCNN-LSTM for fourth-year university students was 18.7% higher than that of 1DCNN-LSTM for fourth-year university students. Compared with 1DCNN-LSTM, 2DCNN-LSTM can achieve better recognition accuracy. The research results can help psychological counselors better understand the emotional state and needs of students and provide more personalized and targeted counseling services.

The purpose of the study is to help users know about their favorite music and expand art market audiences. First, the personalized recommendation data of classical music are obtained based on the deep learning recommendation algorithm technology, artificial intelligence, and music playback software of users. Second, a systematic experiment is conducted on the improved recommendation algorithm, and a classical music dataset is established and used for model training and user testing. Then, the network model of the classical music recommendation algorithm is constructed through the typical convolutional neural network model, and the optimal parameters suitable for the model are found. The experimental results show that the optimal value of the dimension in the hidden layer is 192, and 24,000 training rounds can converge to the global optimum when the learning rate is 0.001. The personalized recommendation is provided for target users by calculating the similarity between user preference and potential features of classical music, relieving the auditory fatigue of art market audiences, improving user experience, and expanding the art market audience through the classical music recommendation system.

Early diagnosis of maize-plant phytosanitary state in the field is crucial to prevent crop damage and optimize yield. However, this field diagnosis presents a challenge due to the variable background of the field environment, which can hinder the performance of classification algorithms. In this article, we introduced a novel segmentation technique using a combined normalized difference vegetation index that effectively isolates the features of interest, such as the leaves, from the surrounding image, which includes the diverse field background. To assess the effectiveness of our segmentation approach, we conducted early diagnosis of maize plants in the field using supervised classification algorithms. We generated a dataset that incorporated four essential texture features: energy, entropy, contrast, and inverse. These features were extracted from each of the segmented images using grayscale co-occurrence matrices. We employed four different classification methods, namely Adaboost, Random Forest, K-Nearest Neighbors, and support vector machine. When combined with the proposed segmentation technique, the support vector machine outperformed the other models, achieving an accuracy rate of 97%.

Despite the advancements in modern computer hardware and software, the creation of digital animation still demands a substantial investment of both manpower and time. This article aimed to explore how artificial intelligence (AI) technology can be combined with traditional animation creation techniques to achieve better integration effects. By combining intelligent character animation generation with hand drawing, a generative adversarial network was used to achieve high-quality animation generation. The generator generated realistic animations, and the discriminator measured the authenticity of the animations by comparing the differences between the generator-generated animations and the real animations, which was used for automated character animation generation. This can greatly reduce the cost and time of digital animation creation, improve the quality of digital animation, and provide more innovation for the application of traditional animation technology. The average number of audience attracted by characters through blended creation was 122.2% more than that of characters through traditional animation creation and 32.6% more than that of characters through AI creation. This not only helps animation producers complete animation production more quickly, but also enhances the creativity and artistic value of animation.

With the widespread popularity of intelligent education, artificial intelligence plays an important role in the field of education. Currently, there are issues such as low accuracy and low adaptability. By comparing algorithms such as logistic regression, decision tree, random forest (RF), support vector machine, and long short-term memory (LSTM) recurrent neural network (RNN), this article adopted a multi-classification fusion strategy and fully considered the adaptability of the algorithm to evaluate and grade students in two scenarios with different grades and teaching quality. By encoding and normalizing student grades, six evaluation parameters were selected for the evaluation criteria of teaching quality through principal component analysis feature selection. Multi-classifier models were used to fuse the five models in pairs, improving the accuracy of the experimental evaluation. Finally, the experimental data of the six fused multi-classification models in the scenarios of student performance estimation and teaching quality estimation were compared, and the experimental effects of education evaluation and grading under different models were analyzed. The experimental results showed that the LSTM RNN-RF model had the strongest adaptability in the scenario of student performance estimation, with an estimation accuracy of 98.5%, which was 12.9% higher than a single RF model. This experiment was closely related to educational scenarios and fully considered the adaptability of different machine learning algorithms to different scenarios, improving the prediction and classification accuracy of the model.

Dance education in colleges and universities is the most important means of inheriting dance skills, cultivating dance talents, and promoting the prosperity and development of dance art. In the new era, the country’s emphasis on “aesthetic education” has provided fertile policy soil for the development of dance majors in universities. Based on the spiritual and cultural needs of the people and the development needs of the national dance art, it is of great urgency for colleges and universities to explore the future oriented Chinese dance higher education and dance creation. Dance education and dance creation are closely linked and interdependent. In the process of Dance education, dance creation inspiration is stimulated. Dance creation and innovation inject new soul into Dance education. College Dance education should combine the two organically to promote the high-quality development of Chinese dance art. The classroom teaching quality evaluation in dance aesthetic education is classical multiple-attributes decision-making (MADM). The probabilistic hesitancy fuzzy sets (PHFSs) are used as a tool for characterizing uncertain information during the classroom teaching quality evaluation in dance aesthetic education. In this paper, we extend the classical grey relational analysis (GRA) method to the probabilistic hesitancy fuzzy MADM with unknown weight information. Firstly, the basic concept, comparative formula and Hamming distance of PHFSs are introduced. Then, the information entropy is used to compute the attribute weights based on the expected values and deviation degree. Then, probabilistic hesitancy fuzzy GRA (PHF-GRA) method is built for MADM under PHFSs. Finally, a practical case study for classroom teaching quality evaluation in dance aesthetic education is designed to validate the proposed method and some comparative studies are also designed to verify the applicability.

Online teaching is not limited by time, but the problem of low learning efficiency is common. To address this problem, the study proposes an attention mechanism for multidimensional feature fusion, which first detects faces, uses a supervised gradient descent algorithm for face feature point detection, and improves the least-squares ellipse-fitting algorithm to detect the open/closed state of human eyes. The sight detection method is also improved, and the fuzzy inference method is used to identify students’ emotions, and the modules are fused to achieve multidimensional feature fusion attention detection for online teaching. The study found that the average accuracy rate was 84.5% with glasses and 92.0% without glasses. The research method with glasses had an average time consumption of 17 ms, while the method without glasses took 15 ms, indicating higher detection accuracy and faster real-time performance. The improved approach led to higher recognition accuracy and accuracy rate. The detection accuracy of a single feature and the research method was 74.1 and 91.9%, respectively. It shows that the research method helps in the detection of students’ attention in online teaching.

Road toll tax contributes significantly in the economic development of any nation. In developing countries, the toll tax collection is carried out either manually or electronically. However, both approaches suffer from various challenges, including prolonged waiting times, lack of transparency, high operational costs, and concerns regarding data security and privacy. This research aims to address these challenges using a blockchain-based system. The proposed system employs advanced image processing techniques, specifically “You Only Look Once” version 5 (YOLOv5), to accurately capture and store vehicles’ registration numbers in a local server situated at toll plazas. Subsequently, the vehicle identification, along with the driver’s credentials, is transmitted to an application server, where an Ethereum smart contract verifies the information and automatically deducts the toll charges from the driver’s account. The results from this study indicate that the proposed system effectively reduces vehicle waiting time and facilitates uninterrupted vehicular movement. Additionally, the system ensures transaction transparency, safeguards the security and privacy of vehicle details, facilitates non-stop payments, rendering unnecessary cash payments or radio-frequency identification scanning at toll booths, and incorporates a decentralized architectural framework to enhance security and mitigate potential system failures.

Recently, the level of criminals and terrorists using light weapons (such as knives and firearms) has increased rapidly around the world. Unfortunately, most current surveillance systems are still based mainly on human monitoring and intervention. For that reason, the requirement for a smart system for detecting different weapons becomes crucial in the field of security and computer vision. In this article, a novel technique for detecting various types of weapons has been proposed. This system is based mainly on deep learning techniques, namely, You Only Look Once, version 8 (YOLOv8), to detect a different class of light weapons. Furthermore, this study focuses on detecting two armed human poses based on ensemble learning techniques, which involve combining the outputs of different Yolov8 models to produce an accurate and robust detection system. The proposed system is evaluated on the self-created weapons dataset comprising thousands of images of different classes of weapons. The experiment results of this work show the effectiveness of ensemble learning for detecting various weapons with high accuracy, achieving 97.2% of mean average precision.

Advances in medical imaging and artificial intelligence have led to improvements in diagnosis and non-invasive patient examination accuracy. The use of the fundamental method for Magnetic resonance imaging (MRI) brain scans as a screening tool has increased in recent years. Numerous studies have proposed a variety of feature extraction methods to classify the abnormal growths in MRI scans. Recently, the MRI texture analysis and the use of deep features have resulted in remarkable performance improvements in the classification and diagnosis of challenging pathologies, like brain tumors. This study proposes employing a handcrafted model based on Weierstrass elliptic function (WEF) and deep feature based on DenseNet-201 to classify brain tumors in MRI images. By calculating the energy of each individual pixel, the Weierstrass coefficients of the WEF are used to capture high frequency image details of the brain image. The WEF mode works to extract the nonlinear patterns in MRI images based on the probability of each pixel. While the dense connectivity of DenseNet-201’s architecture allows to learn features at multiple scales and abstraction levels. These features are passed to support vector machines classifier, which classifies the brain tumor. The results of classification accuracy achieved is 98.55% for combined features of WEF with trained DenseNet-201. Findings on the brain tumor segmentation dataset indicated that the proposed method performed better than alternative techniques for classifying brain tumors.

Geometric and floral models are an important part of clothing and have been used for thousands of years. Although the styles of geometric models and flower models have undergone changes over the centuries, they are still one of the important factors of clothing patterns, the important carrier of aesthetics, and the manifestation of people’s spiritual views and cultural needs. The development and application of digital printing technology have freed people from excessive dependence on sewing and embroidery processes. Therefore, while deeply studying the design of clothing patterns, this work sorted and analyzed the geometric flower models of clothing through interactive genetic algorithms, and optimized programming to enrich the models of clothing with geometric textures. The results showed that the deviation value of geometric flower pattern design was constantly decreasing, while the optimal strategy value was constantly increasing. The mean deviation value was 0.82, which was a decrease of 0.21 on the seventh day compared with the first day; the mean value of the optimal strategy value was 0.84, which was an increase of 0.19 on the seventh day compared with the first day. The visual effect and creativity of the clothing flower pattern design under the interactive genetic algorithm are better than the traditional flower pattern design, and the visual effect and creativity under the interactive genetic algorithm are 9% higher than the traditional one.

At present, blockchain technology has been applied in different fields such as finance, Internet of Things, logistics, digital copyright, and public services, but it is less used in the field of paper-cut animation. Therefore, the use of blockchain to design a copyright protection model for paper-cut animation design based on blockchain is crucial to the progress of paper-cut animation design. This work combines various technologies such as blockchain technology, InterPlanetary File System (IPFS) technology, and Node.js technology for integrated design and construction. In the construction of the system model, the relevant characteristics of the blockchain technology show that the blockchain can provide a new and effective method for digital copyright protection; IPFS technology provides distributed and secure storage for massive amounts of oil and gas knowledge and achievement data; and the relevant development framework and technical characteristics of Node.js technology provide a reliable programming foundation for the realization of the system. Through the construction of the digital copyright protection system model based on the blockchain, the blockchain technology is used to realize the protection of the digital copyright of the knowledge achievement data of paper-cut animation. The results show that the model has high security and is more economical and practical than other existing models. In a word, the copyright protection model of paper-cut animation design based on blockchain can fully protect the rights and interests of creative talents of paper-cut animation, thereby promoting the progress of paper-cut animation industry.

During the artistic journey, creators frequently encounter challenges stemming from pressure, resource constraints, and waning inspiration, all of which can impede their creative flow. Addressing these obstacles requires a multifaceted strategy aimed at nurturing creativity throughout the artistic process. Procedural art generation emerges as a viable solution to invigorate artistic creativity. In this study, the deep Q-network (DQN) was constructed to solve the shortage of artistic creativity through its automatic decision-making ability. The model was trained with different types of artistic styles (abstract and minimalism) in WikiArt dataset. The model generates various artistic elements of different styles, forms, or thinking according to the input parameters or constraints, and selects specific colors, textures, or shapes to help the artist maintain focus in the creation process and expand the creativity in the creation process. In order to achieve this goal, in the process of performing the procedural art generation task with DQN, the experiment collected the generation speed, interpretability, and creativity evaluation feedback of each style of art. The feedback results show that the scores of color field painting and minimalism were 83.2, 93.5, 86.3 and 86.6, 91.5, 82.1 respectively. The research shows that employing dynamic mass spectrometry networks enables the automation of the art creation process. This innovative approach facilitates the exploration of diverse creative ideas tailored to various artistic tasks, thereby fostering advancements in art creation and nurturing creativity.

The traditional machine intelligence system lacks deep understanding and reasoning ability. This study took the automatic driving system in multi-agent as an example to bring higher-level intelligence and decision-making ability to automatic driving through knowledge intelligence. It obtained real-world geographic information data from OpenStreetMap, preprocessed the data, and built a virtual environment. The inception model was used to identify information in environmental images, and the knowledge information of traffic regulations, road signs, and traffic accidents was expressed to build a knowledge map. The knowledge related to automatic driving was integrated, and automatic driving training was carried out through the reward mechanism and the deep Q-network (DQN) model. About 13 kinds of traffic situations were set up in the virtual environment, and the traditional machine intelligence autonomous driving and knowledge fusion autonomous driving multi-agent were compared. The results show that the average number of accidents in 100,000 km of traditional machine intelligence autonomous driving and knowledge fusion autonomous driving multi-agents was 3 and 1.4, and the average number of violations in 100,000 km was 4.3 and 1.8, respectively. The average graphics processing unit (GPU) utilization rate of knowledge fusion autonomous driving in 13 virtual environments was 75.9%, and the average peak GPU utilization rate was 96.1%. Knowledge fusion of autonomous driving multi-agents can effectively improve the safety of autonomous driving and enable autonomous driving multi-agents to have a higher level of decision-making ability.

As the number of online news texts continues to increase, the algorithm of automatic keyword extraction becomes a key content in facilitating users’ fast access to the desired content. This article first introduced two common algorithms: term frequency–inverse document frequency (TF–IDF) and TextRank. Then, the calculation of news title weight was added to the TF–IDF algorithm according to the characteristics of network news text. Moreover, a new automatic extraction algorithm was designed by applying Word2vec to extract semantics. The experimental results demonstrated that on the ACE2005 dataset, as the quantity of automatically extracted keywords increased, the accuracy of the TF–IDF, TextRank, and the semantics-combined TF–IDF algorithms gradually decreased, and the recall rates gradually increased. When five keywords were extracted, the gap of the semantics-combined TF–IDF algorithm with the other two algorithms was the largest, and its accuracy, recall rate, and F -measure were 72.77, 78.64, and 75.59%, respectively. Finally, the F -measure of the semantics-combined TF–IDF algorithm reached 81% for network news texts. The experimental results prove the performance of the semantics-combined TF–IDF algorithm in automatically extracting keywords from network news texts, and it will have promising applications in practice.

The multilayer perceptron (MLP) neural network is a widely adopted feedforward neural network (FNN) utilized for classification and prediction tasks. The effectiveness of MLP greatly hinges on the judicious selection of its weights and biases. Traditionally, gradient-based techniques have been employed to tune these parameters during the learning process. However, such methods are prone to slow convergence and getting trapped in local optima. Predicting urban air quality is of utmost importance to mitigate air pollution in cities and enhance the well-being of residents. The air quality index (AQI) serves as a quantitative tool for assessing the air quality. To address the issue of slow convergence and limited search space exploration, we incorporate an opposite-learning method into the Jaya optimization algorithm called EOL-Jaya-MLP. This innovation allows for more effective exploration of the search space. Our experimentation is conducted using a comprehensive 3-year dataset collected from five air quality monitoring stations. Furthermore, we introduce an external archive strategy, termed EOL-Archive-Jaya, which guides the evolution of the algorithm toward more promising search regions. This strategy saves the best solutions obtained during the optimization process for later use, enhancing the algorithm’s performance. To evaluate the efficacy of the proposed EOL-Jaya-MLP and EOL-Archive-Jaya, we compare them against the original Jaya algorithm and six other popular machine learning techniques. Impressively, the EOL-Jaya-MLP consistently outperforms all other methods in accurately predicting AQI levels. The MLP model’s adaptability to dynamic urban air quality patterns is achieved by selecting appropriate values for weights and biases. This leads to efficacy of our proposed approaches in achieving superior prediction accuracy, robustness, and adaptability to dynamic environmental conditions. In conclusion, our study shows the superiority of the EOL-Jaya-MLP over traditional methods and other machine learning techniques in predicting AQI levels, offering a robust solution for urban air quality prediction. The incorporation of the EOL-Archive-Jaya strategy further enhances the algorithm’s effectiveness, ensuring a more efficient exploration of the search space.

In order to solve the problem of the low transmission rate of wireless network sensors (WNSs), this article proposes a WNS visual assistance system based on embedded systems. The system uses WNSs instead of traditional wired sensors, which can achieve synchronous transmission of multi-node data, enabling nodes to work together better, thereby improving the real-time and reliability of the entire system. This article conducts in-depth research on feature extraction algorithms and tests the visual assistance system in the experimental section. The results show that the recognition rate and stability of the visual symbol assistance system implemented using WNSs are higher than those of ordinary systems. In the satisfaction survey, it was found that 87 people were very satisfied with the visual symbol assistance system, accounting for 87%, while only 57 people were very satisfied with the traditional visual symbol assistance system, accounting for 57%. The experimental results show that the system output stability of the design method is good, and the response time and reliability are better.

With the increasing social pressure and academic competition, the mental health (for convenience, abbreviated as MH) problems of college students are becoming increasingly prominent, but there are often challenges that are difficult to accurately predict and intervene in a timely manner. The aim of this article is to address the early warning needs of college students’ MH problems and construct a model that can timely identify the MH problems of college students. The experiment collected MH related data from college students in S city, and analyzed and trained these data using the Long short-term memory (LSTM) network model. By changing the number of hidden layers, learning rate, batch size, and epoch times, the most suitable training effect was achieved. By using the time-series characteristics of the LSTM model, the selected parameters from the experiment can better capture the changing trends of college students’ MH status, thereby improving prediction accuracy. Finally, three stage indicators of low, medium, and high were set up for early warning of the predicted results, in order to effectively and timely take measures. The research results indicated that the constructed model achieved a minimum regularization loss of 0.0674 after training. Finally, the adjusted model was used to predict the test set, with an average accuracy of 0.852 and an average accuracy of 0.906. The LSTM-based MH risk model performed well in predicting college students’ MH problems and could identify potential risk factors in a timely manner.

Online education resources are more and more abundant, which brings some challenges to learners’ personalized selection. How to provide personalized recommendation services from massive resources according to the needs of learners has gradually become the focus of scholars’ research. Therefore, this article improves the traditional collaborative filtering recommendation algorithm and constructs a personalized recommendation model of an online learning platform based on a long-term memory network and collaborative filtering. First, the stack noise reduction autoencoder combined with auxiliary information is used to extract the user potential vector, and the project potential vector is extracted by using the short-duration memory network and the attention mechanism. Then, the double attribute scoring matrix is used to divide the attributes, and the backpropagation network is used to predict the scores. Through the experimental analysis, the hit rate and recall rate of the model constructed by the research institute are 0.7548 and 0.7247, respectively, and the cumulative gain of normalized loss and running time are 0.3385 and 2.72 s, respectively. This model can effectively make up for the defects of the traditional algorithm caused by cold start and sparse score data and provide more effective and high-quality learning resource recommendations for students.

With the popularization of higher education, the number of students in colleges and universities is increasing, and how to timely cope with the various problems faced by students in employment has become a major problem faced by teachers in colleges and universities. Due to the low utilization rate of student information by the traditional employment management of college graduates, the quality of employment guidance services is not high. Therefore, to solve this problem, this study proposes a simplified, improved Iterative Dichotomizer 3 (ID3) based on the correlation coefficient, and the algorithm improves the information gain function and simplifies the information entropy formula. The experimental results show that the simplified modified ID3 based on correlation coefficients converges faster than the other two algorithms, starting to converge after only 17 iterations; the loss value is also smaller than the other algorithms, at around 0.12. Its minimum accuracy, precision, recall, and F 1-measure for employment status prediction were 86.4, 76.8, 72.8, and 0.82%, respectively, all higher than the rest of the algorithms. The time complexity at a sample size of 80 is only 32 ms, which is lower than the rest of the algorithms. It can be seen that the simplified and improved ID3 based on correlation coefficients can accurately and efficiently perform predictive analysis of graduates’ employment status. The university employment management system proposed in the study has achieved efficient deep utilization of graduate information through ID3, providing assistance to university employment decision-makers and reference for employment guidance for university graduates.

With the development of the Internet, the number of electronic texts has increased rapidly. Automatic grammar error correction technology is an effective safeguard measure for the quality of electronic texts. To improve the quality of electronic text, this study introduces a moving window algorithm and linear interpolation smoothing algorithm to build a Cn-gram language model. On this basis, a syntactic analysis strategy is introduced to construct a syntactic error correction model integrating Cn-gram and syntactic analysis, and English grammar intelligent error correction is carried out through the model. The results show that compared with the Bi-gram and Tri-gram, the precision of the Cn-gram model is 0.85 and 0.91% higher, and the F 1 value is 0.97 and 1.14% higher, respectively. Compared with the results of test set Long, the Cn-gram model has better performance on verb error correction of the Short test set, and the precision rate, recall rate, and F 1 value are increased by 0.86, 3.94, and 1.87%, respectively. The comparison of the precision, recall rate, and F 1 value of the proposed grammar error correction model on the complete test set shows that the precision of the study is 19.10 and 5.41% higher for subject–verb agreement errors. The recall rate is 9.55 and 10.77% higher, respectively; F 1 values are higher by 12.65 and 10.59%, respectively. The above results show that the error-correcting technique of the research design has excellent error-correcting performance. It is hoped that this experiment can provide a reference for the relevant research of automatic error correction technology of electronic text.

The traditional translation robot is limited to the translation of single-mode text images and text videos, which has the problem of low translation accuracy. Therefore, speech recognition and intelligent translation in multimodal human–computer interaction (HCI) system are proposed. First, the network structure of speech recognition model in multi-channel HCI system is established, and the multi-head self-attention mechanism is constructed. Then, the artificial intelligence voice wake-up function is designed, and a multimodal machine translation model is constructed. On this basis, selective attention is added to obtain visual recognition of perceived text, and the decoder is used for multimodal gating fusion to realize the output of encoder translation results. Experimental results show that this method has high BLUE value and high translation accuracy.

Concealing sensitive information inside multimedia is very interesting in secure communication due to its wide application. This study discusses the different difficulties involved in embedding the ciphertext inside video frame without affecting the imperceptibility and the strength resistance against different cyberattacks. The main goal of this study is to suggest a novel technique to conceal encrypted data inside video frames securely. This is done by combining the Laplacian of Gaussian (LoG) edge detection algorithm and Chacha20 encryption algorithm. LoG facilitates to detect the suitable location inside video frame for concealing data. While ChaCha20 is used to encrypt data before embedding them inside these locations, which augmenting data security. The proposed method involves sequence of steps involving detecting edges inside video frames, determining the suitable edges for concealing data, encrypting data by XORing it with the encryption key which is generated using ChaCha20 algorithm, embedding the encrypted data inside the determined edges, and then reconstruct the video frame to rebuild the video that involves the concealed encrypted data and keeping the lowest level of visual distortion. The experimental results showed that combining these two approaches provide fast, robust, and secure method, which can be seen by evaluating the system using mean square error, peak signal-to-noise ratio, correlation, number of pixels change rate, unified average changing intensity, and entropy, these evaluation metrics provide excellent results. This study suggests a strong and novel method to embed the encrypted data inside video frames which can be employed in secure communication, copyright protection, and data authentication. Merging LoG and ChaCha20 algorithms produce perfect results in both security and visual perception quality, that provides a means for farther achievement in secure data embedding techniques.

In response to the limitations of the existing online learning system regarding the efficiency and accuracy of the question-and-answer (Q&A) teacher recommendation method, this research develops a Q&A teacher recommendation model utilizing a Graph Convolutional Neural Network. First, a time-sensitive online learning Q&A teacher recommendation model (A Time Sensitive Online Learning Q&A Teacher Recommendation Model; TSRM) is proposed to address the shortcomings that current recommendation methods ignore, i.e., the teacher’s ability to answer questions with time changes. Then, a TSRM based on Short and Long Term Interest for Answering Questions (LSTR) is proposed to address the problem that the current recommendation methods ignore, i.e., the types of questions of student users’ concerns can change. Finally, we combine the TSRM model and LSTR model to build an intelligent recommendation model for answering teachers. The conclusion is that the accuracy rate of TSRM model on the test set is 99.5%, and the recommendation success rate of LSTR model reaches 98.4%, which are better than the other two models. The above results can show that the LSTR model and TSRM model constructed by the study have high performance and can effectively perform the recommendation of answering teachers in the online learning system, thus improving the efficiency of solving students’ problem, improving students’ learning effect, and contributing to the development of university education informatization.

Remote sensing (RS) image processing has made significant progress in the past few years, but it still faces some problems such as the difficulty in processing large-scale RS image data, difficulty in recognizing complex background, and low accuracy and efficiency of processing. In order to improve the existing problems in RS image processing, this study dealt with ConvNext-convolutional neural network (CNN) and big data (BD) in parallel. Moreover, it combined the existing RS image processing with the high dimensional analysis of data and other technologies. In this process, the parallel processing of large data and high-dimensional data analysis technology improves the difficulty and low efficiency of large-scale RS image data processing in the preprocessing stage. The ConvNext-CNN optimizes the two modules of feature extraction and object detection in RS image processing, which improves the difficult problem of complex background recognition and improves the accuracy of RS image processing. At the same time, the performance of RS image processing technology after neural networks (NNs) and BD fusion and traditional RS image processing technology in many aspects are analyzed by experiments. In this study, traditional RS image processing and RS image processing combined with NN and BD were used to process 2,328 sample datasets. The image processing accuracy and recall rate of traditional RS image processing technology were 81 and 82%, respectively, and the F 1 score was about 0.81 ( F 1 value is the reconciled average of accuracy and recall, a metric that combines accuracy and recall to evaluate the quality of the results, a higher F 1 value indicates a better overall performance of the retrieval system). The accuracy rate and recall rate of RS image processing technology, which integrates NN and BD, were 97 and 98%, respectively, and its F 1 score was about 0.97. After analyzing the process of these experiments and the final output results, it can be determined that the RS image processing technology combined with NN and BD can improve the problems of large-scale data processing difficulty, recognition difficulty under complex background, low processing accuracy and efficiency. In this study, the RS image processing technology combined with NN and BD has stronger adaptability with the help of NN and BD technology, and can adjust parameters and can be applied in more tasks.

The Internet era resulted in the rise and advancement of MOOK, WeChat, and mobile networks, making it possible to expand English teaching methods. However, the English teaching industry has the problem of not valuing students’ personalized cognition, and the accuracy of teaching resource delivery is low. Therefore, the research uses the noise gate analysis method to design a cognitive diagnostic model for students and designs an English teaching resource recommendation model in view of a convolutional joint probability matrix (JPM) decomposition algorithm. The research results showed that the cognitive diagnostic model designed in the study had a higher accuracy. Compared to traditional algorithms, the overall recommendation effect of the English teaching resource recommendation model had an average improvement of 11.63% and compared to the JPM algorithm combined with cognitive diagnosis (CD), the overall recommendation effect value had an average improvement of 1.977%. When recommending complex teaching resources, the recommendation effect value had an average improvement of 11.54% compared to traditional algorithms, and the overall average improvement was 1.877% compared to the JPM algorithm combined with CD. In the experimental group, with the assistance of the research algorithm, students’ grades improved by an average of 2.38 points, which was significantly higher than the 0.89 points in the control group. The experiment showcases that the CD and recommendation model designed by the research has higher accuracy, can help improve the efficiency of teaching resource recommendation, reduces teaching costs, and has certain application value.

State-of-the-art segmentation models have shown great performance in structured road segmentation. However, these models are not suitable for the wild roads, which are highly unstructured. To tackle the problem of real-time semantic segmentation of wild roads, we propose a Multi-Information Concatenate Network based on BiSeNetV2 and construct a segmentation dataset Dalle Molle institute for artificial intelligence feature segmentation (IDSIAFS) based on Dalle Molle institute for artificial intelligence. The proposed model removes structural redundancy and optimizes the semantic branch based on BiSeNetV2. Moreover, the Dual-Path Semantic Inference Layer (TPSIL) reduces computation by designing the channel dimension of the semantic branch feature map and aggregates feature maps of different depths. Finally, the segmentation results are achieved by fusing both shallow detail information and deep semantic information. Experiments on the IDSIAFS dataset demonstrate that our proposed model achieves an 89.5% Intersection over Union. The comparative experiments on Cityscapes and India driving dataset benchmarks show that proposed model achieves good inference accuracy and faster inference speed.

A significant component of the online learning platform is the online exercise assessment system, which has access to a wealth of past student exercise data that may be used for data mining research. However, the data from the present online exercise system is not efficiently used, making each exercise less relevant for students and decreasing their interest and interaction with the teacher as she explains the activities. In light of this, this research creates an exercise knowledge map based on the connections between workouts, knowledge points, and previous tournaments. The neural matrix was then improved using cross-feature sharing and feature augmentation units to deconstruct the workout recommendation model. The study also developed an interactive text sentiment analysis model based on the expansion of the self-associative word association network to assess how students interacted after the introduction of the personalized exercise advice teaching approach. The outcomes demonstrated that the suggested model’s mean diversity value at completion was 0.93, an increase of 0.14 and 0.23 over collaborative filtering algorithm and DeepFM (deep factor decompose modle), respectively, and that the proposed model’s final convergence value was 92.3%, an improvement of 2.3 and 4.1% over the latter two models. The extended model used in the study outperformed the support vector machine (SVM) and Random Forest models in terms of accuracy by 5.9 and 1.7%, respectively. In terms of F 1 value indicator, the model proposed by the research has a value of 90.4%, which is 2.5 and 2.1% higher than the SVM model and Random Forest model; in terms of recall rate indicators, the model proposed by the research institute has a value of 94.3%, which is an increase of 6.2 and 9.8% compared to the latter two models. This suggests that the study’s methodology has some application potential and is advantageous in terms of customized recommendation and interactive sentiment recognition.

A large number of digital painting image resources cannot be directly converted into electronic form due to their differences in painting techniques and poor preservation of paintings. Moreover, the difficulty of extracting classification features can also lead to the consumption of human time and misclassification problems. The aim of this research is to address the challenges of converting various digital painting image resources into electronic form and the difficulties of accurately extracting classification features. The goal is to improve the usefulness and accuracy of painting image classification. Converting various digital painting image resources directly into electronic format and accurately extracting classification features are challenging due to differences in painting techniques and painting preservation, as well as the complexity of accurately extracting classification features. Overcoming these adjustments and improving the classification of painting features with the help of artificial intelligence (AI) techniques is crucial. The existing classification methods have good applications in different fields. But their research on painting classification is relatively limited. In order to better manage the painting system, advanced intelligent algorithms need to be introduced for corresponding work, such as feature recognition, image classification, etc. Through these studies, unlabeled classification of massive painting images can be carried out, while guiding future research directions. This study proposes an image classification model based on AI stroke features, which utilizes edge detection and grayscale image feature extraction to extract stroke features; and the convolutional neural network (CNN) and support vector machine are introduced into image classification, and an improved LeNet-5 CNN is proposed to achieve comprehensive assurance of image feature extraction. Considering the diversity of painting image features, the study combines color features with stroke features, and uses weighted K-means clustering algorithm to extract sample features. The experiment illustrates that the K-CNN hybrid model proposed in the study achieved an accuracy of 94.37% in extracting image information, which is higher than 78.24, 85.69, and 86.78% of C4.5, K-Nearest Neighbor (KNN), and Bi directional Long short-term Memory (BiLSTM) algorithms. In terms of image classification information recognition, the algorithms with better performance from good to poor are: the mixed model > BiLSTM > KNN > C4.5 model, with corresponding accuracy values of 0.938, 0.897, 0.872, and 0.851, respectively. And the number of fluctuation nodes in the mixed model is relatively small. And the sample search time is significantly shorter than other comparison algorithms, with a maximum recognition accuracy of 92.64% for the style, content, color, texture, and direction features of the image, which can effectively recognize the contrast and discrimination of the image. This method effectively provides a new technical means and research direction for digitizing image information.

Market planning for regional economic development is a crucial task connected to stability and systematicness and influences the unique outcomes of regional economic growth. As a result, market planning for regional economic development must be given more weight. There are several issues with regional economies currently developing in many parts of our country. Lack of focus on market planning for regional economic growth results in several flaws in market planning, making it challenging to support sustainable development. The accurate analysis of diverse massive economic datasets has significant research value. The aim of this study is to provide an overview of market planning and regional economic and industrial development, analyze the effects of market planning on regional economic and industrial development, identify issues with market planning and regional economic and industrial development, and suggest solutions. This article employs big data analysis technology to forecast corporate market value and develop regional economic markets. The system is employed to conduct experimental big data analysis within a specific field for an extended period. According to test analysis results, the system’s data analysis accuracy can reach up to 95.8%, which has a promising future for use.

An enhanced whale optimization algorithm is introduced to address the challenges associated with the logistics distribution center location problem, aiming to overcome the traditional heuristic algorithm’s susceptibility to local optima and enhance the overall efficiency of the logistics system. This improvement involves the integration of stochastic sinusoidal inertia weights and a variational strategy to augment the global search capability and convergence accuracy of the conventional algorithm. The results of experimental simulations indicate that the upgraded whale algorithm exhibits superior computational performance compared to other heuristic algorithms. Furthermore, it demonstrates the ability to effectively compute optimal distribution center locations, thereby enhancing the delivery efficiency of the logistics distribution system.

With the continuous advancement of agricultural Internet of Things (IoT) technologies, real-time monitoring of agricultural environments has become increasingly significant. This monitoring provides valuable information on pest and disease occurrences and corresponding ecological conditions. However, as the agricultural environments have extensive coverage, the number of monitoring IoT devices and the volume of information will rapidly increase. This leads to a surge in network traffic and computing demands. To address this issue, this article proposes an agricultural environmental monitoring IoT system that utilizes edge computing and deep learning technologies. It combines the Long-Range Wide Area Network (LoRaWAN) for long-range transmission with pest recognition and counting modules. By offloading workloads traditionally processed in the cloud to edge nodes, the proposed system effectively reduces transmission and cloud computing pressures for the agricultural monitoring IoT. Simulation experiments demonstrate stable LoRaWAN protocol-based data transmission at the edge, with an overall packet loss rate of less than 5%, meeting the transmission quality requirements. Moreover, this article investigates a pest recognition and counting method based on deep learning technology. Pest images, captured by monitoring nodes, are recognized and counted online using the TensorFlow framework. Experimental results indicate an accuracy of 89% in pest recognition. By digitally transmitting pest image recognition results to the cloud, the proposed system significantly alleviates transmission and cloud computing pressures for the monitoring IoT.

At present, there is a poor connection between theory and practice in the driving mechanism of industry–teaching integration in colleges and universities. And with the increasing courses, the recommendation accuracy of the recommendation algorithm has also decreased. Therefore, the research built a teaching platform of the Internet of Things (IoT) based on the integration of industry and education and improved its internal online education course recommendation algorithm. Meanwhile, experiments verified its performance. The experimental results show that the response time of several important interfaces is maintained between 0 and 300 ms. In the verification experiment of the improved algorithm for building the rule engine, when the rules are 50, the traditional Rete algorithm takes the most time. In terms of total time consumption, the traditional Rete algorithm takes more time than the improved Rete algorithm. The mean absolute error of the User-Characteristics and Interest Clustering (CCIC) algorithm is 0.8116, the root mean square error is 0.9455, the accuracy is 0.3043, and the recall is 0.1475, which are better than the comparison algorithms. In the recommendation of actual agricultural courses, the overall satisfaction of the User-CCIC algorithm is more than 70%, with good prediction accuracy. In general, the IoT education platform based on the combination of industry and education established by this research has better application prospects, and the User-CCIC algorithm recommended by this research has a good practical effect in actual course recommendation.

Analyzing online learning behavior helps to understand students’ progress, difficulties, and needs during the learning process, making it easier for teachers to provide timely feedback and personalized guidance. However, the classroom behavior (CB) of online teaching is complex and variable, and relying on traditional classroom supervision methods, teachers find it difficult to comprehensively pay attention to the learning behavior of each student. In this regard, a dual stream network was designed to capture and analyze CB by integrating AlphaPose human keypoint detection method and image data method. The experimental results show that when the learning rate of the model parameters is set to 0.001, the accuracy of the model is as high as 92.3%. When the batch size is 8, the accuracy of the model is as high as 90.8%. The accuracy of the fusion model in capturing upright sitting behavior reached 97.3%, but the accuracy in capturing hand raising behavior decreased to only 74.8%. The fusion model performs well in terms of accuracy and recall, with recall rates of 88.3, 86.2, and 85.1% for capturing standing up, raising hands, and sitting upright behaviors, respectively. And the maximum F 1 value is 0.931. The dual stream network effectively integrates the advantages of two types of data, improves the performance of behavior capture, and improves the robustness of the algorithm. The successful application of the model is beneficial for teachers’ classroom observation and research activities, providing a favorable path for their professional development, and thereby improving the overall teaching quality of teachers.

The intelligent university teaching detection system can effectively improve the teaching efficiency of university classrooms and improve the teaching environment. In order to improve the quality of online teaching supervision, this study combines digital monitoring technology to monitor the online teaching process in colleges and universities, and analyzes the method and process of using the equivalence line and the rotation theorem to estimate the projection angle. This work applies a projection method to digital teaching detection, improves the algorithm based on requirements, uses factor improvement methods to improve the problems in model application, uses requirement analysis methods to construct a system model, and uses experimental analysis methods to verify the effectiveness of the proposed algorithm and model. With the support of simulation platforms and mathematical statistical methods, the effectiveness of the system model is verified in this work. From the evaluation results, it can be seen that the evaluation results of the system model in this work are distributed between [84,92], indicating that the use of the system in this study is very excellent. The experimental research shows that the digital online monitoring technology proposed in this study can monitor students in real time in online teaching in colleges and universities, effectively improve the participation of students in the online teaching process, and play an important role in improving the quality of teaching.

Due to insufficient information and feature extraction in existing face-detection methods, as well as limited computing power, designing high-precision and efficient face-detection algorithms is an open challenge. Based on this, we propose an improved face detection algorithm. First, through 1 × 1’s common convolution block (CBL) expands the channel for feature extraction, introduces a depthwise separable residual network into the YOLO-v4 network to further reduce the amount of model computation, and uses CBL to reduce the dimension, so as to improve the efficiency of the subsequent network. Second, the improved attention mechanism is used to splice the high-level features, and the high-level features and the shallow features are fused to obtain the feature vectors containing more information, so as to improve the richness and representativeness of the feature vectors. Finally, the experimental results show that compared with other comparative methods, our method achieves the best results on public face datasets, and our performance in personal face detection is significantly better than other methods.

With the rapid development of China’s economy and society, the income level of Chinese residents is increasing day by day, and the dietary consumption structure of domestic residents is constantly optimized and upgraded. The consumption demand for livestock products such as meat, eggs and milk, and corn starch industrial products is increasing day by day. Therefore, as the main raw material of feed industry and deep processing owners, the consumption scale of China corn market is constantly expanding. However, the corresponding domestic corn production scale has not expanded, and the domestic corn production presents an obvious problem of insufficient supply scale, which leads to a sharp and rapid increase in the number of corn imports in China, which seriously impacts the domestic corn market and affects China’s healthy development of corn industry. In recent years, the continuous spread of the COVID-19 epidemic has seriously affected the global grain production and trade. The fierce military conflict between Russia and Ukraine, the two major grain exporting countries, has once again brought a serious adverse impact on the supply and price of the global grain market, resulting in an increase in the import cost of corn in China and a change in the import pattern of corn in China. In this regard, this study analyzes the current situation of corn production and import trade in China, and based on the relevant trade data between China and major corn importing countries from 2013 to 2022, establishes a trade gravity model to quantitatively evaluate the main factors affecting China’s corn import trade, and analyzes the trade potential of sample countries. It was found that in 2022, China’s import and export trade volume and trade value were 20.62 million tons and 7.11 billion US dollars, respectively, compared with 2013, trade volume and trade volume increased by 6.3 times. Statistics on the potential types of various trading countries showed that the proportion of potential countries was 14.2%, and Brazil had the greatest potential in China’s corn import trade. The proportion of potential expanding countries is about 71.4%, and Laos is a mature country with less opportunities for further development. In order to promote the benign development of the domestic corn industry and the development of China’s corn import trade, this paper puts forward a series of countermeasure suggestions based on the above research.

In order to solve the problems that the current English grammar correction algorithms are not effective, the error correction ability is limited, and the error correction accuracy needs to be improved, this study proposes an automatic grammar correction method for business English writing based on two-way long short-term memory (LSTM) and N-gram. First, this study considers article and preposition errors as a special sequence labeling task, and proposes a Grammar error checking (GEC) method for sequence labeling based on bidirectional LSTM. During training, english as a second language (ESL) corpus and supplementary corpus are used to label specific articles or prepositions. Second, for noun simple-plural errors, verb form errors, and subject-verb inconsistency errors, a large number of news corpora are used to count the frequency of N-gram, and a GEC method based on ESL and news corpora N-gram voting strategy is proposed. Experimental results show that the overall F 1 value of the method designed in this study on the GEC data of CoNLL2013 is 33.87%, which is higher than the F 1 value of UIUC. The F 1 value of article error correction is 38.05%, and the F 1 value of preposition error correction is 28.89%. It is proved that this method can effectively improve the accuracy of grammar error correction and solve the gradient explosion problem of traditional error correction model, which is of great significance to further strengthen the practicality of automatic grammar error correction technology.

In the field of tourism recommendation systems, accurately recommending scenic spots and routes for users is one of the hot research directions. In order to better consider the complex interaction between user preferences and attraction features, as well as the potential connections between different information sources, this study constructed a graph attention network model using knowledge graphs for tourist attraction and route recommendations, and extracted features from visual images using visual geometry group-16. The results indicate that, in Xian, when the learning rate is 0.01, the area under the curve value is 0.916. The area under the curve of New York is 0.909, and the learning rate is 0.001. The area under the curve value of the Tokyo dataset is 0.895. When the learning rate is moderate, the model quickly stabilizes in the first 16 rounds and reaches its optimal state in 26–30 rounds. When the propagation depth is 2, the accuracy is 0.920, 0.905, and 0.895, respectively. After introducing visual features, the accuracy, recall, and F 1 score improved by 10 to 15.7%. The multi-layer perceptron further increased the effect by 4–6%. These experimental data fully demonstrate the effectiveness and accuracy of the recommendation algorithm. This study provides a powerful tool for tourism recommendation systems, which helps to further improve user experience.

In the investment risk assessment of green finance (GF) projects, the application of multi-attribute perceptual fuzzy information decision technology is taken as the main research object. With the promotion of the concept of environmental protection and the development of green economy, the investment risk assessment of GF projects has become more and more important. However, this requires dealing with a large amount of fuzzy information and multi-attribute decision problems, which is a big challenge for traditional decision techniques. Based on this background, a new decision model, intuitionistic fuzzy preference theory-based tomada de decisão interativa multicritério (IF-PT-TODIM), is adopted, which can better deal with fuzzy information and multi-attribute decision problems by taking two different choices as reference. By knowing the weight distribution of experts, the model can better assess the influence of various factors on the decision. In the research results, the calculated results of expert weights are 0.2796, 0.2221, 0.1914, 0.1328, and 0.1745, respectively, showing that each expert has different degrees of influence on decision-making. In addition, the application of IF-PT-TODIM model can effectively reduce the investment risk. Compared with national bank of Kuwait, systematic review, evolutionary algorithm, the improved method can reduce the risk of venture capital by 28.14, 15.47, and 11.05%, respectively. This result further confirms the advantage of the IF-PT-TODIM model in dealing with fuzzy information and multi-attribute decision problems. This study has practical implications for understanding and improving the investment risk assessment of GF projects. It not only provides a new decision model for risk assessment, but also provides an effective method to deal with fuzzy information and multi-attribute decision problems. This provides new ideas and methods for the risk management of GF projects and also provides a new perspective and reference for research in related fields.

At present, the amount of information on power grid operation and maintenance monitoring image data is increasing, and the requirements for data compression are higher and higher. Based on the improved SPIHT image compression algorithm, this study presents the research of power grid data compression. First, the basic theory of image compression, and the principle of one-dimensional wavelet transform and two-dimensional wavelet transform are introduced. The development process, characteristics, and advantages of image coding are discussed. Then, the shortcomings of the SPIHT algorithm are analyzed, and the SPIHT coding is improved by parallel computation. The parallel wavelet transform algorithm based on the block idea and the parallel SPIHT coding algorithm based on the code tree are proposed in the data parallelism of the compression algorithm. At the same time, the data dependence between tasks in the process of SPIHT image compression coding is analyzed, and the task parallelism in the compression algorithm is realized by using the relative independence of tasks in different threshold coding. Finally, the application and simulation analysis of power grid data based on the SPIHT compression algorithm, the construction of power grid data model simulation, and the composition of two-dimensional power grid data images are carried out. Secondly, the obtained 2D power grid data image is compressed by the SPIHT algorithm and improved SPIHT algorithm, respectively, and the compression effect of the two algorithms on the power grid data image is compared. When the bit rate is 0.5, the compression effect of the improved SPIHT algorithm is 13.6506. When the bit rate is 1, the compression effect of the improved SPIHT algorithm is 18.9287. The results show that the improved SPIHT algorithm can compress the grid data to obtain better grid image quality.

Electric vehicles (EVs) as a sustainable safety system are being increasingly used and receiving attention from researchers for several reasons including optimal performance, affordability for consumers, and environmental safety. EV speed control is a crucial issue that requires reliable and intelligent controllers for maintaining this matter. The primary goal of this research is to design the linear and nonlinear Proportional, Integral, and Derivative (PID) controllers to control EV speed based on the minimum value of ITAE plus ISU (integral square of control signal) as well as satisfy the constrain on response overshoot. All the proposed PID controllers, conventional PID controller, arc tan PID controller, and nonlinear PID controller (NL-PID) are used in cascade with EV model. In all these PID controllers, a filter is used with the derivative term to avoid the effect of the noise. The tuning of the proposed controller gains is achieved using Aquila Optimization algorithm. The controllers’ parameter tuning is primarily determined by reducing the Integral Time Absolute Error (ITAE) and integral square control signal. Numerical simulation, system modelling, and controller design are done using MATLAB. By comparing the results, the proposed controllers’ efficacy is demonstrated. The proposed NL-PID controller provides promising EV speed regulation control and robustness to external disturbances. Where, the performance specifications of the proposed NL-PID controller when using unit step input with step disturbance of 0.2 and −0.2, and the model parameters increased by 25% from its nominal value are the rise time is 2.608, settling time is 2.608, overshoot is 0.073%, the maximum control signal is 6.230, number of slope sign change in the control signal is 231, and ITAE is 17.6064. The comparative results show the NL-PID controller’s superior performance for tracking the reference signal with the lowest peak for the control signal, rejection of disturbances, ability to overcome model uncertainties, and lowest ITAE value.

In the field of traditional landscape painting and art image restoration, traditional restoration methods have gradually revealed limitations with the development of society and technological progress. In order to enhance the restoration effects of Chinese landscape paintings, an innovative image restoration algorithm is designed in this research, combining edge restoration with generative adversarial networks (GANs). Simultaneously, a novel image restoration model with embedded multi-scale attention dilated convolution is proposed to enhance the modeling capability for details and textures in landscape paintings. To better preserve the structural features of artistic images, a structural information-guided art image restoration model is introduced. The introduction of adversarial networks into the repair model can improve the repair effect. The art image repair model adds a multi-scale attention mechanism to handle more complex works of art. The research results show that the image detection model improves by 0.20, 0.07, and 0.06 in the Spearman rank correlation coefficient, Pearson correlation coefficient, and peak signal-to-noise ratio (PSNR), respectively, compared to other models. The proposed method outperforms mean filtering, wavelet denoising, and median filtering algorithms by 6.3, 9.1, and 15.8 dB in PSNR and by 0.06, 0.12, and 0.11 in structural similarity index. In the image restoration task, the structural similarity and information entropy indicators of the research model increase by approximately 9.3 and 3%, respectively. The image restoration method proposed in this study is beneficial for preserving and restoring precious cultural heritage, especially traditional Chinese landscape paintings, providing new technological means for cultural relic restoration.

A traceability and analysis method for measurement laboratory testing data based on the intelligent Internet of Things (IoT) and deep belief network (DBN) is proposed to address the issue of low accuracy in identifying anomalies in measurement testing data and difficulty in identifying the causes of anomalies. First, a data analysis system for the metrology laboratory is designed based on an intelligent IoT architecture of “cloud-management-edge-end.” Then, the Gaussian Bernoulli-Restricted Boltzmann machine is introduced to improve the DBN model, which is deployed on the edge side for learning the ledger data sample library to determine the anomaly detection data of the metrology device. Finally, a stacked denoising autoencoder model is used in the cloud center to extract historical electricity consumption curve features, and the cause of anomalies is determined by calculating the cosine similarity between it and the target device feature curve to complete traceability analysis. Based on the selected dataset, the proposed method is experimentally demonstrated, and the results show that its traceability accuracy and time consumption are 88.72% and 3.949 s, respectively, which can meet the detection requirements of the metrology laboratory.

Problem In recent years, computing and sensing advances have helped to develop efficient human posture classification systems, which assist in creating health systems that contribute to enhancing elder’s and disable’s life quality in context-aware and ambient assistive living applications. Other applications of body posture classification include the simulation of human bodies into virtual characters, security applications such as kidnapping and the body position of the kidnapper or victim, which can provide useful information to the negotiator or the rescue team, and other sports applications. Aim This work aims to propose a body posture classification system based on speech using deep learning techniques. Methods All samples pass through the preprocessing phase. The Mel-frequency cepstral coefficient (MFCC) with 12 coefficients was used as features, and the best features were selected by the correlation-based feature selection method. Two supervised learning techniques called artificial hydrocarbon networks (AHN) and convolutional neural networks (CNN) were deployed to efficiently classify body postures and movements. This research aims to detect six human positions through speech, which include walking downstairs, sitting, walking upstairs, running, laying, and walking. The dataset used in this work was prepared by the authors, named the human activity speech dataset. Results The best performance gained was with the AHN classifier of 68.9% accuracy and 67.1% accuracy with the CNN. Conclusion It was concluded that the MFCC features are the most powerful features in body position classification, and the deep learning methods are powerful in classifying body positions.

The deep fusion of Internet technology and education is constantly pushing forward the reform of university education. Traditional educational ideas, concepts, and models cannot keep pace with the times, and hybrid teaching has become a new way of education in colleges and universities. To improve the teaching effect of physical education classes, the study used a blended teaching model and designed a teaching evaluation and performance prediction model under the blended teaching model based on an improved cluster analysis method and attention mechanism. The lab results indicated that under the blended teaching model, students’ performance increased by 12.89 points, and the level of skill mastery and proficiency increased by 26.52 and 28.55%, respectively, with grades more inclined to high score distribution. “Excellent” grade clustering increased by 77.71%, and “Good” grade clustering increased by 19.01%. The minimum error sum of squares of the improved clustering algorithm was 58.18 and 36.25% lower than the other two algorithms, and the clustering results were more relevant. The two-way attention mechanism algorithm predicted higher accuracy results and performed best on all four evaluation metrics, with a prediction accuracy of 98.23%, an accuracy of 98.42%, and an F 1 value of 91.78%. This hybrid teaching model is more in line with the characteristics of the physical education teaching discipline, successfully cultivates students’ independent learning ability, stimulates students’ love for physical education courses, and achieves better teaching results.

Resource-constrained problems for technology-based applications/services are common due to pervasive utilization and in-definite user/demand densities. Traditional resource allocation methods consume high allocation time and make it difficult to predict the possible solutions from the collection of resources. Various range of solutions through optimizations are provided for addressing the issues that, however, result in imbalanced solutions. This article assimilates genetic algorithm (GA) and fuzzy clustering process and introduces resource-constrained reduction framework. The proposed framework utilizes a GA for mutating the allocation and availability possibilities of the resources for different problems. The possibilities of solutions are tailored across various demands preventing replications. Post this process, the fuzzy clustering segregates the optimal, sub-optimal, and non-optimal solutions based on the mutation rate from the genetic process. This reduces the complexity of handling heterogeneous resources for varying demand, user, and problem densities. Based on the clustering process, the crossover features are tailored across multiple resource allocation instances that mitigate the existing constraints. This proposed framework improves the problem-addressing ability (11.44%) and improves resource allocation (8.08%), constraint mitigation (11.1%), and allocation time (11.85%).

In the process of behavior recognition, the recognition operation may be carried out in various environments such as sunny, cloudy, and night. Since traditional recognition algorithms are judged by identifying the pixels of the image, the intensity of the light will affect the image. The brightness and contrast of the display thus interfere with the recognition results. Therefore, traditional algorithms are easily affected by the lighting environment around the recognition object. To improve the accuracy and recognition rate of the behavior recognition algorithm in different lighting environments, a convolutional neural network (CNN) algorithm using a dual-stream method of time flow and spatial flow is studied here. First, we collect behavioral action data sets and preprocess the data. The core of the behavior recognition algorithm of the dual-stream residual CNN is to use the time stream and the spatial stream to fuse behavioral features and eliminate meaningless data features. After processing Perform feature selection on the data, select the acoustic wave and light-sensing features of the data, and finally, use the extracted features to classify and identify using the two-stream residual CNN and the traditional behavior recognition method. The behavior recognition algorithm based on the dual-stream residual CNN was tested on the data of four groups of people. For the behavioral feature map with a data volume of 50, the behavior recognition algorithm of the dual-stream residual CNN was effective in various environments under different lighting conditions. The recognition accuracy can reach 83.5%, which is 12.3% higher than the traditional. The behavior recognition algorithm of the dual-stream residual CNN takes 17.25 s less than the conventional recognition algorithm. It is concluded that behavior recognition based on dual-stream residual CNNs can indeed improve the recognition accuracy and recognition speed in environments with different lighting conditions than traditional behavior recognition.

One of the key factors driving a country’s economic development and ensuring the sustainability of its industries is the constant availability of electricity. This is normally provided by the national grid. However, the power supply is not always stable in developing countries where new businesses, including the telecommunications industry, are constantly emerging. Therefore, they must rely on generators to ensure their full functionality. These generators rely on fuel to function, and consumption is usually high if not properly monitored. Monitoring is usually done by a (non-expert) human. This can sometimes be a tedious process, as some companies have reported excessively high consumption rates. For anomaly detection in power generating plants, the studies by Mulongo et al. and Atemkeng and Jimoh used the same dataset to train a multilayer perceptron (MLP) and generative adversarial networks (GANs), respectively, achieving an accuracy of 96.1% with MLP and 98.9% with GAN. Through comparative analysis and the use of ensemble learning techniques, we found that ensemble learning models outperform both MLP and GAN as proposed by Mulongo et al. and Atemkeng and Jimoh using the same dataset. Furthermore, we investigated the potential of autoencoders to outperform MLPs, GANs, and ensemble learning models. To this end, we have introduced a label-assisted autoencoder approach for detecting anomalies in power-generating plants. This model includes a labelling assistance module that adjusts the thresholds. Our results indicate that the label-assisted autoencoder outperforms the MLP. However, GANs and all ensemble learning models outperformed the label-assisted autoencoder. Nevertheless, the use of a label-assisted autoencoder offers a distinct advantage in categorizing anomalies based on their severity, a capability not present in ensemble learning models and GANs.

As the information technology develops, educational resource recommendation system has become an indispensable part of the education field. At the same time, people’s demand for personalised educational resources is also growing; therefore, how to accurately predict user demand and provide personalised recommendations has become an important issue. In this study, a fuzzy predictive control model on the grounds of neural network is proposed to optimise the design of educational resource recommendation in the context of information technology. After experimental testing, the model recommendation fit reached 95.16 and 92.91% on the two test datasets, which are significantly higher than the control model. The average F 1 values of the proposed model also reached 95.21 and 88.77%, which are higher than the control model. In other control experiments, the proposed model of this study also has a better performance. The relevant outcomes showcase that the predictive performance and recommendation effect of the model can be further enhanced by improving the structure of the neural network and the parameter optimisation method. Meanwhile, the proposed model has high performance in information overload and personalised demand, which offers a useful reference for the optimal design of educational resource recommendation system.

The medical laboratory plays a crucial role within a hospital setting and is responsible for the examination and analysis of patient specimens to accurately diagnose various ailments. The burden on medical laboratory personnel has significantly increased, particularly in the context of the ongoing global COVID-19 pandemic. Worldwide, the implementation of comprehensive and extended COVID-19 screening programs has placed a significant strain on healthcare professionals. This burden has led to exhaustion among medical employees, limiting their ability to effectively track laboratory resources, such as medical equipment and consumables. Therefore, this study proposed an artificial intelligence (AI)-based solution that contributes to a more efficient and less labor-intensive workflow for medical workers in laboratory settings. With the ultimate goal to reduce the burden on healthcare providers by streamlining the process of monitoring and managing these resources, the objective of this study is to design and develop an AI-based system for consumables tracking in medical laboratories. In this work, the effectiveness of two object detection models, namely, YOLOv5x6 and YOLOv8l, for the administration of consumables in medical laboratories was evaluated and analyzed. A total of 570 photographs were used to create the dataset, capturing the objects in a variety of settings. The findings indicate that both detection models demonstrate a notable capability to achieve a high mean average precision. This underscores the effectiveness of computer vision in the context of consumable goods detection scenarios and provides a reference for the application of real-time detection models in tracking systems within medical laboratories.

As society develops and educational needs continue to change, the traditional way of teaching ideology and politics is facing challenges in terms of efficiency and effectiveness evaluation. In response to the low efficiency of ideological and political education (IPE) methods and the difficulty in accurately and comprehensively evaluating students’ ideological and political literacy and moral qualities, this article used the Long Short-Term Memory with Self-Attention Mechanism (LSTM-SAM) model to conduct experiments on the evaluation of IPE effectiveness. First, by collecting information on IPE from a research center of a certain university in 2023, and then using the LSTM (Long Short-Term Memory) model to catch the long-term dependencies of IPE, the learning trajectory and changing trends of students can be better understood. The self-attention mechanism was applied to dynamically learn and distinguish the importance of different parts in the input sequence, better weighting key features such as student learning behavior and participation level, thereby enhancing the accuracy and robustness of effectiveness evaluation. Finally, the splicing method was adopted to integrate the LSTM model and self-attention mechanism for the experiment, and the teaching efficiency of different teaching methods was statistically analyzed through a questionnaire survey. The test results indicated that the classification accuracy of the LSTM-SAM model reached 98.41%, which was 1.61% higher than the LSTM model. The teaching efficiency was the highest under the gamified teaching method, providing an effective method for evaluating the effectiveness of IPE and providing useful reference for optimizing teaching methods.

In recent years, tourist attractions have become increasingly popular vacation destinations, leading to a gradual emergence of the tourism market and an improvement in the regulatory systems of scenic spots. However, these attractions frequently experience various abnormal behaviors (ABs). The existing abnormal behavior detection (ABD) algorithms are hindered by interference from the scenic area background, resulting in poor identification of ABs. The study proposed a background model constructed by an optimized Gaussian mixture model based on the background subtraction method to eliminate the background interference. Based on this model, an ABD method was established using action data based on spatio-temporal block detection and motion foreground effect map features. Experiments were conducted to train the video clips and action databases to establish the ABD model. In the study, algorithms were tested using computers, and different scenic spots were set up to validate tourists. Three scenic spot scenes with different time and environmental conditions were set up to detect tourist ABD, and the results were compared with different existing anomaly detection algorithms. Research has shown that the area under the receiver operation characteristic curve in outdoor scenes a and c is significantly larger than that in indoor scene b, with areas under the curves of 96.55 and 97.40%, respectively. In two outdoor research scenarios b and c, the area under the curve and equal error rate of the research algorithm (RA) are 88.14, 18.24, 98.12, and 7.55%, respectively, which are significantly better than the two compared algorithms. Although they are 97.31 and 6.29% in scenario a, slightly lower than the 95.18 and 12.69% of the compared algorithms, the expected time is very close to the ideal value, and overall, they have good performance. In addition, the average recognition accuracy of the RA (93.24%) is higher than that of the comparative algorithm (89.32%). The proposed ABD method has shown certain efficiency and accuracy in identifying ABs. It can provide effective decision-making support for scenic area management and has reference significance for improving the safety management quality of scenic area monitoring. By monitoring the AB of tourists, scenic area management personnel can take timely measures to ensure the safety of tourists and the order of the scenic area. In addition, the results of this study not only provide a new approach and method for detecting AB of scenic spot tourists, but also provide valuable reference significance and theoretical contributions in fields such as video surveillance, behavior analysis, and artificial intelligence.

One major difficulty facing the healthcare industry is the early and precise detection of melanoma. With its capacity to record a broad spectrum of electromagnetic wavelengths, hyperspectral imaging (HSI) is a promising technique for accurate melanoma diagnosis. However, the limited accessibility of HSI technology prevents it from being widely used. This article introduces a novel method for converting readily available red green blue (RGB) images to their hyperspectral counterparts using convolutional neural networks (CNNs). Through this conversion process, spectral information is improved, enabling a more thorough examination to detect melanoma. Principal component analysis (PCA) aids the machine learning algorithm in differentiating between melanoma and healthy moles in the classification model. Accuracy is significantly increased when spot detection and PCA are combined; Naïve Bayes achieves 76% accuracy in this way. These models are used in the developed web-based program SkinScan for real-time melanoma analysis, providing a useful and accessible solution. This work emphasizes how CNN-driven RGB-to-HSI conversion might improve melanoma detection accuracy and accessibility.

In highway bridge engineering, it is essential to use modern image processing methods to effectively detect and classify road bumps and potholes, considering the unique characteristics of bridge surfaces. Accurate identification and handling of road surface irregularities are crucial for preserving the longevity and security of transportation infrastructure. The study proposes a Convolutional Neural Network for Highway Anomaly Detection (BCNN-HAD) that uses the “BridgeGuard-Vision” (BGV) method, a computer vision technology for highway bridges to increase the accuracy and efficiency of automated image processing for more accurate detection of road irregularities in highway surveillance. Through model training using datasets that imitate various environmental conditions frequently seen on highway bridges, the proposed approach obtains dependable characteristics from photos taken close to bridges, improving flexibility and accuracy. The training method considers variations in lighting, weather conditions, and bridge materials, ensuring the model performs well in various real-world situations. In addition, this work explores the combination of sensor fusion techniques, combining data from many sources such as bridge structural health monitoring systems, cameras, accelerometers, and Global Positioning System. This comprehensive method, represented by BGV-YOLOv5, aims to offer a complete understanding of the bridge surroundings, therefore helping to detect road irregularities and further developing the field of bridge health monitoring. Expected results involve developing a personalized and effective system for detecting road bumps and potholes to tackle specific difficulties in highway bridge situations. In addition, the project seeks to provide a structure for smart transportation systems in the field of bridge engineering. This project seeks to address the particular requirements of highway bridge engineers, improving road safety and infrastructure maintenance methods in highway bridge engineering with the overall objective of establishing a safer and longer-lasting transportation network.

Identification and classification of bugs, e.g., security and performance are a preemptive and fundamental practice which contributes to the development of secure and efficient software. Software Quality Assurance (SQA) needs to classify bugs into relevant categories, e.g., security and performance bugs since one type of bug may have a higher preference over another, thus facilitating software evolution and maintenance. In addition to classification, it would be ideal for the SQA manager to prioritize security and performance bugs based on the level of perseverance, severity, or impact to assign relevant developers whose expertise is aligned with the identification of such bugs, thus facilitating triaging. The aim of this research is to compare and analyze the prediction accuracy of machine learning algorithms, i.e., Artificial neural network (ANN), Support vector machine (SVM), Naïve Bayes (NB), Decision tree (DT), Logistic regression (LR), and K-nearest neighbor (KNN) to identify security and performance bugs from the bug repository. We first label the existing dataset from the Bugzilla repository with the help of a software security expert to train the algorithms. Our research type is explanatory, and our research method is controlled experimentation, in which the independent variable is prediction accuracy and the dependent variables are ANN, SVM, NB, DT, LR, and KNN. First, we applied preprocessing, Term Frequency-Inverse Document Frequency feature extraction methods, and then applied classification algorithms. The results were measured through accuracy, precision, recall, and F -measure and then the results were compared and validated through the ten-fold cross-validation technique. Comparative analysis reveals that two algorithms (SVM and LR) perform better in terms of precision (0.99) for performance bugs and three algorithms (SVM, ANN, and LR) perform better in terms of F 1 score for security bugs as compared to other classification algorithms which are essentially due to the linear dataset and extensive number of features in the dataset.

In this work, we propose a novel cross-dataset micro-expression identification method based on facial regions of interest contribution quantification, where the training samples are from the source dataset and the test samples are from the target dataset. Specifically, a micro-expression video clip (i.e., a sample) is first sampled at intervals to capture an image sequence consisting of multiple consecutive video frames. Then, locate and crop the facial area in the image frame, and perform facial landmark detection on the cropped area. Next calculate the optical-flow field of the image sequence, and extract the improved Main Directional Mean Optical-flow feature. Finally, the learned group sparse model is used as the classifier to predict the label information of the unmarked target samples, so as to complete the category recognition of micro-expressions. Extensive cross-dataset comparative experiments on four spontaneous micro-expression datasets, CASME, CASME II, SMIC-HS, and SAMM, show that the recognition strategy proposed in this work is effective. Compared with several other advanced recognition methods, our method has better cross-dataset recognition performance, not only the classification accuracy is higher, but also the classification stability is better when faced with different target datasets and different micro-expression categories. It can be applied in many fields such as clinical diagnosis, interrogation, negotiation, and national security.

Multiple sclerosis (MS) is a dangerous illness that strikes the central nervous system. The body’s immune system attacks myelin (an entity above the nerves) and impairs brain-to-body communication. To date, it is not possible to cure MS. However, symptoms can be managed, and treatments can be provided if the disease is diagnosed early. Hence, supervised machine learning (ML) algorithms and several hyperparameter tuning techniques, including Bayesian optimization, have been utilized in this study to predict MS in patients. Descriptive and inferential statistical analysis has been conducted before training the classifiers. The most essential markers were chosen using a technique called mutual information. Among the search techniques, the Bayesian optimization search technique prevailed to be pre-eminent, with an accuracy of 89%. To comprehend the diagnosis generated by the ML classifiers, four techniques of explainable artificial intelligence were utilized. According to them, the crucial attributes are periventricular magnetic resonance imaging (MRI), infratentorial MRI, oligoclonal bands, spinal cord MRI, breastfeeding, varicella disease, and initial symptoms. The models could be deployed in various medical facilities to detect MS in patients. The doctors could also use this framework to get a second opinion regarding the diagnosis.

The electromagnetic-gravity optimization (EMGO) framework is a novel optimization technique that integrates the fine-structure constant and leverages electromagnetism and gravity principles to achieve efficient and robust optimization solutions. Through comprehensive performance evaluation and comparative analyses against state-of-the-art optimization techniques, EMGO demonstrates superior convergence speed and solution quality. Its unique balance between exploration and exploitation, enabled by the interplay of electromagnetic and gravity forces, makes it a powerful tool for finding optimal or near-optimal solutions in complex problem landscapes. The research contributes by introducing EMGO as a promising optimization approach with diverse applications in engineering, decision support systems, machine learning, data mining, and financial optimization. EMGO’s potential to revolutionize optimization methodologies, handle real-world problems effectively, and balance global exploration and local exploitation establishes its significance. Future research opportunities include exploring adaptive mechanisms, hybrid approaches, handling high-dimensional problems, and integrating machine learning techniques to enhance its capabilities further. EMGO gives a novel approach to optimization, and its efficacy, advantages, and potential for extensive adoption open new paths for advancing optimization in many scientific, engineering, and real-world domains.

In the dynamic landscape of China’s booming economy, the surge in e-commerce customer volume presents both opportunities and challenges, notably in managing customer churn (CC). Addressing this critical issue, this study introduces an innovative approach employing a radial basis function neural network for predicting CC within the e-commerce sector. To enhance the model’s performance in handling the vast and complex data inherent to e-commerce, the least absolute shrinkage and selection operator regression algorithm is employed, optimizing the model’s predictive accuracy. By meticulously analyzing the customer lifecycle, this refined model adeptly predicts churn at various stages, enabling the identification of features most correlated with churn. Empirical results underscore the model’s exceptional capability, achieving a prediction accuracy of 95% and a remarkably low loss rate of 3%. Furthermore, during the excavation, advanced, stable, and decline stages of the customer lifecycle, accuracy levels of 97.6, 93.1, 92.7, and 91.8% are attained, respectively, facilitating the precise selection of highly correlated customer features. Thus, the advanced churn prediction model proposed herein significantly contributes to the e-commerce domain, offering a robust tool for strategizing customer retention and mitigating churn.

Sub-health problems are becoming increasingly serious in today’s society, and some organizations are not paying enough attention to adolescent sports health data. For adolescent sports, health needs to be measured regularly and tested constantly so that the intake of diet and medication can be reasonably adjusted according to their biochemical indicators. The Smart Health Life Growth Cloud System can effectively manage residents’ health data digitally and informally, enabling users to manage their health data better and facilitating doctors to keep abreast of users’ health conditions, while also facilitating the government to conduct research and studies on the physical fitness of adolescents in the areas under its jurisdiction. The cloud-based management platform for student physical health management relies on the mobile internet as a practical service platform whose primary role is to provide young people with a convenient sporting life, focusing on practicality, service, and interactivity. We also collect sensor data to detect gait patterns (with or without leg contact) and filter them through an adaptive hybrid filter to differentiate between the two patterns. In turn, the Smart Health Life Growth Cloud system changes the traditional medical model and greatly improves the information and intelligence of the healthcare industry. Using the exercise individual health evaluation model in this article is controlled to be within 20%, thus concluding that the exercise individual health evaluation model proposed in this article can predict the exercise limit of an exercise individual more accurately.

Audio–visual speech synthesis (AVSS) is a rapidly growing field in the paradigm of audio–visual learning, involving the conversion of one person’s speech into the audio–visual stream of another while preserving the speech content. AVSS comprises two primary components: voice conversion (VC), which alters the vocal characteristics from the source speaker to the target speaker, followed by audio–visual synthesis, which creates the audio–visual presentation of the converted VC output for the target speaker. Despite the progress in deep learning (DL) technologies, DL models in AVSS have received limited attention in existing literature. Therefore, this article presents a novel approach for AVSS utilizing capsule network (Caps-Net)-based autoencoders, with the incorporation of cycle consistency loss. Caps-Net addresses translation invariance issues in convolutional neural network approaches for effective feature capture. Additionally, the inclusion of cycle consistency loss ensures the retention of content information from the source speaker. The proposed approach is referred to as cycle consistency loss-based capsule autoencoders for audio–visual speech synthesis (CCLCap-AE-AVSS). The proposed CCLCap-AE-AVSS is trained and tested using VoxCeleb2 and LRS3-TED datasets. The subjective and objective assessments of the generated samples demonstrate the superior performance of the proposed work compared to the current state-of-the-art models.

With the recent growth of global populations, main roads in cities have witnessed an evident increase in the number of vehicles. This has led to unprecedented challenges for authorities in managing the traffic of ambulance vehicles to provide medical services in emergency cases. Despite the high technologies associated with medical tracks and advanced traffic management systems, there is still a current delay in ambulances’ attendance in times of emergency to provide patients with vital aid. Therefore, it is indispensable to introduce a new emergency service system that enables the ambulance to reach the patient in the least congested and shortest paths. However, designing an efficient algorithm to plan the best route for an ambulance is still a global goal and a challenge that needs to be solved. This article introduces an Internet of Things emergency services system based on a real-time node rank index (NR-index) algorithm to find the best route for the ambulance to reach the patient and provide the required medical services in emergency cases. The proposed system design copes with the dynamic traffic conditions to guarantee the shortest transport time. For this purpose, a vehicular ad hoc network is employed to collect accurate real-time traffic data. In this article, we suggest two parameters to compromise distance and congestion level. The first is the distance between the patient and the surrounding ambulance vehicles, and the second determines the congestion level to avoid the path with high congestion traffic. The system employs a developed real-time NR-index algorithm to select a suitable ambulance vehicle to respond to emergency cases at a low travel cost with the fastest journey. Finally, our system makes it easier for ambulance vehicles to use the best route and avoid heavy traffic. This allows them to make their way to the patient quickly and increases the chance of saving lives. The simulation results show significant improvements in terms of average travel time, average travel speed, and normalized routing load.

Problem Corneal topography instruments have limited parameter constraints for calculating precise defect ratios on the basis of the cone base area of the anterior axial curvature map for patients with Keratoconus (KC). Aim The aim of this study is to use thresholding-based segmentation and morphological techniques to calculate the pathological ratio of the keratoconic cornea through cone base area extraction for the detection of KC severity and tracking of disease development. Methods Data were collected from February 2022 to March 2023, comprising 97 cases from private clinics in southern Iraq. Disease severity was categorized into three stages, namely, mild, moderate, and severe, according to the topographic KC classification by a senior ophthalmologist. The Galilei system was used in obtaining the corneal topography images. The study proposed an image analysis method for corneal topography images using MATLAB R2020a. The method had four main steps: preprocessing, image segmentation, morphological processing, and pathological ratio calculation. Moreover, pathological ratio was compared with the KC severity through statistical analysis. A P -value less than 0.05 indicated statistically significant results. Results The majority of the cases in the mild category had a pathological ratio of ≤20%, and the moderate category had a higher prevalence ranging from 21 to 40%. The severe category had the highest distribution (<40%). A P -value of <0.001 indicated significant and clear link between KC stages and pathologic ratio. Conclusion The algorithm used for extracting the cone base area of the keratoconic cornea at different stages was validated by an ophthalmic specialist to ensure that the cone base area was appropriately extracted. The findings may help ophthalmologists to make informed decisions for patients with severe KC and assessments based on the percentage of corneal defects.

Due to the limitations of the initial rapidly exploring random tree (RRT) algorithm, robotics faces challenges in path planning. This study proposes the integration of the metaheuristic salp swarm algorithm (SSA) to enhance the RRT algorithm, resulting in a new algorithm termed IRRT-SSA. The IRRT-SSA addresses issues inherent in the original RRT, enhancing efficiency and path-finding capabilities. A detailed explanation of IRRT-SSA is provided, emphasizing its distinctions from the core RRT. Comprehensive insights into parameterization and algorithmic processes contribute to a thorough understanding of its implementation. Comparative analysis demonstrates the superior performance of IRRT-SSA over the basic RRT, showing improvements of approximately 49, 54, and 54% in average path length, number of nodes, and number of iterations, respectively. This signifies the enhanced effectiveness of the proposed method. Theoretical and practical implications of IRRT-SSA are highlighted, particularly its influence on practical robotic applications, serving as an exemplar of tangible benefits.

Limited device resources and an ever-changing cybersecurity landscape compound the challenges faced by the network protection infrastructure for Internet of medical things (IoMT) applications, which include different device ecosystems, privacy concerns, and problems with interoperability. Protecting private medical information in IoMT apps is challenging; a comprehensive strategy that provides user education, standard protocols, and robust security mechanisms is necessary to overcome these obstacles. With the advancement of IoMT, the network of clinical systems, gadgets, and sensors is integrated with the Internet of things (IoT) to enable intelligent healthcare solutions. However, the sensitive data sharing and the substantial connections in the IoMT systems raise security and privacy concerns in the network. Therefore, network security is critical in IoMT applications due to data breaches, vulnerabilities, and distributed denial of service attacks on medical data. This study reviews the network security techniques implemented in the existing studies for IoMT applications using machine learning and blockchain technology. This study presents an overview of IoMT healthcare applications by highlighting the security challenges encountered and the necessity of adopting advanced techniques to deal with complex threats. The research is mainly about how deep reinforcement learning (DRL), commonly used for intrusion detection, access control, and anomaly detection, works over time and how it can be used in IoMT applications. With the notion of providing robust security in IoMT applications, this study appraises the benefits of blockchain technology, such as data integrity, accountability, and confidentiality. Besides, this study addresses the limitations and challenges of various security techniques that IoMT systems employ. This work assesses the findings, research gaps, and future advancements for enhancing network security in IoMT applications. With an extensive analysis of existing research, this survey guides researchers, medical practitioners, and decision-makers to scale up the DRL and blockchain in IoMT systems more efficiently in the future.

Network security faces increasing threats from denial of service (DoS) and distributed denial of service (DDoS) attacks. The current solutions have not been able to predict and mitigate these threats with enough accuracy. A novel and effective solution for predicting DoS and DDoS attacks in network security scenarios is presented in this work by employing an effective model, called CNN-LSTM-XGBoost, which is an innovative hybrid approach designed for intrusion detection in network security. The system is applied and analyzed to three datasets: CICIDS-001, CIC-IDS2017, and CIC-IDS2018. We preprocess the data by removing null and duplicate data, handling imbalanced data, and selecting the most relevant features using correlation-based feature selection. The system is evaluated using accuracy, precision, F 1 score, and recall. The system achieves a higher accuracy of 98.3% for CICIDS-001, 99.2% for CICIDS2017, and 99.3% for CIC-ID2018, compared to other existing algorithms. The system also reduces the overfitting of the model using the most important features. This study shows that the proposed system is an effective and efficient solution for network attack detection and classification.

These days, 5G mobile technology highly influences the wireless communication process, which is more responsive than the other generation. However, the 5G communication process handles the speed and latency issues while transmitting data over the network. Especially in rural areas, most people face this speed and latency problem while connecting to the network. Therefore, the network uses massive amounts of data to analyse this problem by integrating 5G with the Internet of things (IoT). The IoT-based data transmission and collection helps manage the data capacity, connection, and transmission rate. Therefore, this study uses the novel secured narrowband IoT framework for 5G communication and operational goals based on their position in the market. The 5G network uses business connectivity and access information to manage the data transmission. The network uses cellular bandwidth and IoT devices to connect more during this process. According to the connection, the information is transmitted with the minimum delay. Thus, experimental results have proposed these parameters: an accuracy ratio of 92.5%, a security ratio of 94.4%, a reliability ratio of 93.5%, a latency ratio of 91.8%, a cost-effective ratio of 20.5%, and a throughput ratio of 25.4% compared to other methods.

Problem Metaheuristics are efficient algorithms designed to address a broad spectrum of optimization challenges and offer satisfactory solutions, even in scenarios of limited processing capability or incomplete information. It has been observed that no single metaheuristic algorithm is universally ideal for all applications. This realization underscores the opportunity for the introduction of new metaheuristic algorithms or enhancements to existing ones. Aim The aim of this work is to propose Quokka swarm optimization (QSO), a novel nature-inspired metaheuristic optimization technique. The QSO simulates the cooperative behavior of quokka animals, which can be used to address optimization issues. Method A group of common unconstrained and constrained test functions is employed to demonstrate the strength of the proposed approach. To test the performance of QSO, 43 popular test functions that are used in the optimization were employed as benchmarks. The solutions have been refining their positions in tandem with the ongoing discovery of the best solution. In addition, QSO can substitute the worst quokka with the best child found so far to improve the solutions. Performance comparisons using the Blue monkey swarm optimization, Gray wolf optimization, Biogeography-based optimizer, Artificial bee colony, Particle swarm optimization, and Gravitational search algorithm were also performed. Results The obtained results showed that QSO is competitive in comparison to the chosen metaheuristic algorithms.

In the face of escalating cyber threats, a real-time automated security evidence collection system for cloud-based digital forensics investigations is essential for identifying and mitigating malicious activities. However, the substantial volumes of data generated by modern cloud-based digital systems pose difficulties in collecting and analyzing evidence promptly and systematically. To address these challenges, this research introduces an architecture that combines a security lake and a modern data lake. The primary objective of this architecture is to overcome the obstacles associated with gathering evidence from multiple cloud-based accounts and regions while ensuring the flexibility and scalability required to manage the ever-expanding data volumes encountered in cloud-based digital forensics investigations. This work focuses on gathering security events from multiple accounts and regions within a cloud environment in real-time while maintaining the integrity of the evidence and storing them in lakes, providing investigators with the flexibility to move between these lakes for analysis to get quick results. This is achieved through the utilization of security lake and modern data architecture. To validate the system, we tested it within a university system comprising numerous accounts spread across different regions within an AWS environment. Overall, the proposed system effectively gathers evidence from various sources and consolidates all data lakes into a single account. These lakes were then utilized for analyzing the evidence using Athena and Wazuh.

With the continuous development of computer vision technology, the image modeling of environmental design has also received more and more attention. This study proposed an image modeling algorithm for environment design based on augmented reality (AR) and virtual reality (VR) technologies. On the system architecture, the environment design image modeling algorithm based on AR and VR technologies has provided users with an immersive 3D simulation space environment. It also has provided users with a good interactive interface through AR technology, enabling users to interact with computers in a realistic virtual three-dimensional environment and interactive operation. This study verified the algorithm from three aspects: time, authenticity, and interactivity. When facing object A, the modeling time values of the traditional environmental design image modeling method and the environmental design image modeling method proposed in this study were 12.14 and 8.72 s, respectively. The modeling authenticity reached 87.51 and 98.57%, respectively, and the interactivity reached 79.14 and 96.45%, respectively. Some columns of data have proved that the environmental design image modeling method proposed in this study was superior to the traditional environmental design image modeling method. In addition, this study also verified the system model from three aspects: stability, running smoothness, and performance loss. When the number of image modeling in the environment design was 2, the stability, running smoothness, and performance loss of the system were 99.37, 99.25 and 0.75%, respectively. The experimental data again proved that the algorithm system model proposed in this study was feasible and worthy of further application.

The Internet of Things (IoT) has expanded into a realm of cutting-edge integrated solutions across numerous applications, following three stages of development: communication, authentication, and computing. However, each layer inside the three tier IoT architecture faces a spectrum of security concerns due to the IoT’s openness, scope, and resource limits. Therefore, developing a secure IoT system is vital, shielding against attackers leveraging IoT devices to obtain network access and maintaining data security during transmission between these devices. Despite developments in Intrusion Detection Systems, identifying Denial-of-Service (DoS) and Distributed Denial-of-Service (DDoS) attacks in real-time remains a significant issue, especially in the environment of IoT devices, which are diverse and resource restricted. This study provides a Convolutional Neural Network-Support Vector Machine (CNN-SVM) hybrid technique, targeted at overcoming these limits by enhancing detection speed and accuracy, thus addressing this substantial gap in the area. This research offers a hybrid model that combines CNN for feature extraction with SVM as a classifier. This model employs a CNN to extract essential elements from the data and classifies attacks as either DDoS or benign. Our results highlight the potential performance of our model, which was trained on the real Canadian institute for cybersecurity (CIC) IoT dataset in 2023. Notably, our hybrid model outperformed classic machine learning (ML) alternatives such as CNN, SVM, K-nearest neighbors, Naïve Bayes, and Logistic Regression. During testing, our model attained a remarkable accuracy rate of 99% and an F 1-score of 99%, outperforming the highest-performing SVM ML model with an accuracy of 98% and the other ML methods.

Objective With the popularity of high-resolution devices such as high-definition, ultra-high-definition televisions, and smartphones, the demand for high-resolution images is also increasing, which puts forward higher requirements for high-resolution image processing and entity recognition technology. Method This article introduced the research progress and application of high-resolution image processing and entity recognition algorithms from the perspective of artificial intelligence (AI). First, the important role of AI in high-resolution image processing and entity recognition was introduced, and then the applications of deep learning-based algorithms in high-resolution image grayscale equalization, denoising, and deblurring were introduced. Subsequently, the application of AI-based object detection and image segmentation algorithms in entity recognition was explored, and the superiority of AI-based high-resolution image processing and entity recognition algorithms was verified through training and testing. The accuracy of the model was verified through testing experiments. Finally, a summary and outlook were made on high-resolution image processing and entity recognition algorithms based on AI. Result After experimental testing, it was found that high-resolution image processing and entity recognition based on AI had higher efficiency, and the overall image recognition ability was improved by 29.6% compared to traditional image recognition models. The recognition speed and accuracy were also improved. Conclusion High-resolution image processing and element recognition algorithms based on AI enabled observers to see the detailed information in the image more clearly, thus improving the efficiency and accuracy of image analysis. Through continuous improvement of algorithm performance, real-time application, and expansion of cross-disciplinary applications, people can look forward to the development of more advanced and powerful image processing and entity recognition technologies, which will bring huge impetus to research and application in various fields.

In the realm of image-based breast cancer detection and severity assessment, this study delves into the revolutionary potential of sophisticated artificial intelligence (AI) techniques. By investigating image processing, machine learning (ML), and deep learning (DL), the research illuminates their combined impact on transforming breast cancer diagnosis. This integration offers insights into early identification and precise characterization of cancers. With a foundation in 125 research articles, this article presents a comprehensive overview of the current state of image-based breast cancer detection. Synthesizing the transformative role of AI, including image processing, ML, and DL, the review explores how these technologies collectively reshape the landscape of breast cancer diagnosis and severity assessment. An essential aspect highlighted is the synergy between advanced image processing methods and ML algorithms. This combination facilitates the automated examination of medical images, which is crucial for detecting minute anomalies indicative of breast cancer. The utilization of complex neural networks for feature extraction and pattern recognition in DL models further enhances diagnostic precision. Beyond diagnostic improvements, the abstract underscores the substantial influence of AI-driven methods on breast cancer treatment. The integration of AI not only increases diagnostic precision but also opens avenues for individualized treatment planning, marking a paradigm shift toward personalized medicine in breast cancer care. However, challenges persist, with issues related to data quality and interpretability requiring continued research efforts. Looking forward, the abstract envisions future directions for breast cancer identification and diagnosis, emphasizing the adoption of explainable AI techniques and global collaboration for data sharing. These initiatives promise to propel the field into a new era characterized by enhanced efficiency and precision in breast cancer care.

This study aims to perform a thorough systematic review investigating and synthesizing existing research on defense strategies and methodologies in adversarial attacks using machine learning (ML) and deep learning methods. A methodology was conducted to guarantee a thorough literature analysis of the studies using sources such as ScienceDirect, Scopus, IEEE Xplore, and Web of Science. A question was shaped to retrieve articles published from 2019 to April 2024, which ultimately produced a total of 704 papers. A rigorous screening, deduplication, and matching of the inclusion and exclusion criteria were followed, and hence 42 studies were included in the quantitative synthesis. The considered papers were categorized into a coherent and systematic classification including three categories: security enhancement techniques, adversarial attack strategies and defense mechanisms, and innovative security mechanisms and solutions. In this article, we have presented a systematic and comprehensive analysis of earlier studies and opened the door to potential future studies by discussing in depth four challenges and motivations of adversarial attacks, while three recommendations have been discussed. A systematic science mapping analysis was also performed to reorganize and summarize the results of studies to address the issues of trustworthiness. Moreover, this research covers a large variety of network and cybersecurity applications of defense in adversarial attack subjects, including intrusion detection systems, anomaly detection, ML-based defenses, and cryptographic techniques. The relevant conclusions well demonstrate what have achieved in defense mechanisms against adversarial attacks. In addition, the analysis revealed a few emerging tendencies and deficiencies in the area to be remedied through better and more dependable mitigation methods against advanced persistent threats. The findings of this review have crucial implications for the community of researchers, practitioners, and policy makers in network and cybersecurity using artificial intelligence applications.

The integration of artificial intelligence (AI) and big data technologies has the potential to revolutionize various industries, yet there are complexities and challenges associated with their implementation. This comprehensive study aims to investigate the combined impact of AI and big data on operational efficiency, precision, and security across multiple sectors. By utilizing a methodological analysis of 105 peer-reviewed articles sourced from reputable databases, we systematically explore the diverse applications, key innovations, and transformative potential of these technologies. Our findings uncover significant advancements in healthcare diagnostics, drug discovery, personalized education, and smart farming, highlighting how AI enhances big data analytics to drive notable improvements. Specifically, the study reveals the accuracy of AI in healthcare diagnostics, the efficiency of big data in drug discovery, the personalization of learning experiences through AI in education, and the sustainability advancements in agriculture through smart farming. These results underscore a substantial shift toward more sophisticated data-driven decision-making and operational processes facilitated by the integration of AI and big data. This shift addresses the initial research problem and makes a significant contribution to both academic and practical understanding of the role these technologies play in shaping the future of industry operations. The study concludes that while AI and big data integration offers substantial benefits, addressing associated challenges is crucial for maximizing their impact.

In recent years, the field of data analytics has witnessed a surge in innovative techniques to handle the ever-increasing volume and complexity of data. Among these, nature-inspired algorithms have gained significant attention due to their ability to efficiently mimic natural processes and solve intricate problems. One such algorithm, the symbiotic organisms search (SOS) Algorithm, has emerged as a promising approach for clustering and predictive analytics tasks, drawing inspiration from the symbiotic relationships observed in biological ecosystems. Metaheuristics such as the SOS have been frequently employed in clustering to discover suitable solutions for complicated issues. Despite the numerous research works on clustering and SOS-based predictive techniques, there have been minimal secondary investigations in the field. The aim of this study is to fill this gap by performing a systematic literature review (SLR) on SOS-based clustering models focusing on various aspects, including the adopted clustering approach, feature selection approach, and hybridized algorithms combining K-means algorithm with different SOS algorithms. This review aims to guide researchers to better understand the issues and challenges in this area. The study assesses the unique articles published in journals and conferences over the last ten years (2014–2023). After the abstract and full-text eligibility analysis, a limited number of articles were considered for this SLR. The findings show that various SOS methods were adapted as clustering and feature selection methods in which CSOS, discrete SOS, and multiagent SOS are mostly used for the clustering applications, and binary SOS, binary SOS with S-shaped transfer functions, and BSOSVT are used for feature selection problems. The findings also revealed that, of all the selected studies for this review, only a few studies specifically focused on hybridizing SOS with K-means algorithm for automatic data clustering application. Finally, the study analyzes the study gaps and the research prospects for SOS-based clustering methods.

The metaverse, a virtual multiuser environment, has garnered global attention for its potential to offer deeply immersive and participatory experiences. As this technology matures, it is evolving in tandem with emerging innovations such as Web 3.0, Blockchain, nonfungible tokens, and cryptocurrencies like Bitcoin, which play pivotal roles in the metaverse economy. Robust Bitcoin networks must be modelled for the metaverse environment in Industry 5.0 platforms to ensure the metaverse’s sustained growth and relevance. Industry 5.0 is poised to experience significant economic expansion, driven in large part by the transformative influence of metaverse technology. Researchers have actively explored diverse strategies and approaches to address the unique challenges and opportunities presented by current Bitcoin networks, highlighting the limitless potential for enhancing anonymity and privacy while navigating this exciting digital frontier. By addressing the diverse anonymity and privacy evaluation attributes, the lack of clarity regarding the prioritisation of these attributes and the variability in data, this modelling approach can be categorised as a form of multiple attribute decision-making (MADM). This review seeks to achieve three main objectives: firstly, to identify research gaps, obstacles, and problems within scholarly literature, which is crucial for assessing and modelling Bitcoin networks to succour the metaverse environment of Industry 5.0; secondly, to pinpoint theoretical gaps, proposed solutions, and benchmarking of Bitcoin networks; and thirdly, to offer an overview of the existing validation and evaluation methods employed in the literature. This review introduced a unique taxonomy by intersecting “Bitcoin networks based on blockchain aspects” with “anonymity and privacy development attributes aspect.” It emphasised the study’s significance and innovation. The results illustrate that employing MADM techniques is highly suitable for modelling Bitcoin networks to support the metaverse within the context of Industry 5.0. This thorough review is an invaluable resource for academics and decision-makers, offering perspectives regarding the improvements, applications, and potential directions for evaluating Bitcoin networks to bolster the metaverse environment of Industry 5.0.

Machine learning (ML) and deep learning (DL) techniques have demonstrated significant potential in the development of effective intrusion detection systems. This study presents a systematic review of the utilization of ML, DL, optimization algorithms, and datasets in intrusion detection research from 2018 to 2023. We devised a comprehensive search strategy to identify relevant studies from scientific databases. After screening 393 papers meeting the inclusion criteria, we extracted and analyzed key information using bibliometric analysis techniques. The findings reveal increasing publication trends in this research domain and identify frequently used algorithms, with convolutional neural networks, support vector machines, decision trees, and genetic algorithms emerging as the top methods. The review also discusses the challenges and limitations of current techniques, providing a structured synthesis of the state-of-the-art to guide future intrusion detection research.