Band 1, Heft 4

Forest growth simulation models are useful in evaluating the effects of management practices and climate changes in terrestrial ecosystems, however their successful application requires accurate calibration of model parameters. We have implemented here a stepwise line search (SLS), Gibbs sampling (GS) and preclustering based strength Pareto algorithm (K-SPEA2) to find an optimal set of parameters.

In this paper, we investigate an evolutionary approach to solve the multi-objective dynamic facility layout problem (FLP) under uncertainty that presents the layout as a set of Pareto-optimal solutions. Research examining the dynamic FLP usually assumes that data for each time period are deterministic and known with certainty. However, production uncertainty is one of the most challenging aspects in today’s manufacturing environments. Researchers have only recently modeled FLPs with uncertainty. Unfortunately, most solution methodologies developed to date for both static and dynamic FLPs under uncertainty focus on optimizing just a single objective. To the best of our knowledge, the use of Pareto-optimality in multi-objective dynamic FLPs under uncertainty has not yet been studied. In addition, the approach proposed in this paper is tested using a backward pass heuristic to determine its effectiveness in optimizing multiple objectives. Results show that our approach is an efficient evolutionary dynamic FLP approach to optimize multiple objectives simultaneously under uncertainty.

GPUSVM (Graphic Processing Unit Support Vector Machine) is a Computing Unified Device Architecture (CUDA) based Support Vector Machine (SVM) package. It is designed to offer an end-user a fully functional and user friendly SVM tool which utilizes the power of GPUs. The core package includes an efficient cross validation tool, a fast training tool and a predicting tool. In this article, we first introduce the background theory of how we build our parallel SVM solver using CUDA programming model. Then we compare our GPUSVM package with the popular state of the art Libsvm package on several well known datasets. The preliminary results have shown one to two orders of magnitude speed improvement in both training and predicting phases compared to Libsvm using our Tesla server.

Ubiquitous convergent telecommunication and Internet networks need to deliver services and content to users in a way that is tailored to the users’ context information and preferences. Thus, there is a new challenge in the areas of accurate service description, linking dynamic service discovery and invocation, which involves the services of telecom networks and the Internet. Semantic annotation can provide advantages for precise service description, discovery and composition. However, open service interface specifications of telecommunication networks are currently still in the syntactic level. By applying semantics to Parlay-based services for telecommunication networks and the Internet, we present an OWL-S-based semantic description approach for telecommunication network services, enabled by the telecommunication service domain ontologies to address the semantic interoperability that exists. Using this approach, we have implemented the semantic telecommunication services gateway and proposed the unified service integration architecture of telecommunication and Internet networks within the semantic-web service environment. Proof-of-concept prototype and case studies demonstrate the practical feasibility of the suggested solution.

Most existing performance evaluation methods concentrate on defining various metrics over a wide range of conditions and generating standard benchmarking video sequences to examine the effectiveness of a video tracking system. It is a common practice to incorporate a robustness margin or factor into the system/algorithm design. However, these methods, deterministic approaches, often lead to overdesign, thus increasing costs, or underdesign, causing frequent system failures. In order to overcome the aforementioned limitations, we propose an alternative framework to analyze the physics of the failure process via the concept of reliability. In comparison with existing approaches where system performance is evaluated based on a given benchmarking sequence, the advantage of our proposed framework lies in that a unified and statistical index is used to evaluate the performance of an automated video surveillance system independent of input sequences. Meanwhile, based on our proposed framework, the uncertainty problem of a failure process caused by the system’s complexity, imprecise measurements of the relevant physical constants and variables, and the indeterminate nature of future events can be addressed accordingly.

Ubiquitous computing environments advocate creating applications by plugging together various resources (mobile devices, displays, augmented everyday objects, and so on) and Web Services to support the user’s everyday activities and needs. This approach is referred to as application composition, and such applications are called composite. Due to the dynamic nature of ubiquitous environments, application composition has to be supported at runtime, so that the applications are able to adapt to the situation within the environment and other contexts. Application composition is usually performed by autonomous context-aware mechanisms that provide limited or no control for users. Still, users need to be aware of their environments and be able to control and configure applications when they are composed and executed. Towards this goal we present in this article a context-aware application composition system based on the iCompose interface for composing and controlling applications at runtime. Users compose applications by simply touching the resources in the environment with their mobile devices, while the iCompose interface provides feedback and assists users by suggesting them possible further actions. The interface relies on a rule-based reasoner and utilizes various context sources to support users dynamically, according to the situation in which they compose applications. We present a complete implementation of the system and report the results of a user evaluation study conducted with 21 participants. This study assesses the issues of control, usability, feasibility and user acceptance of the iCompose interface for context-aware application composition and the prototype as a whole.

Density estimation is one of the most important problems in statistical inference and machine learning. A common approach to the problem is to use histograms, i.e., piecewise constant densities. Histograms are flexible and can adapt to any density given enough bins. However, due to the simplicity of histograms, a large number of parameters and a large sample size might be needed for learning an accurate density, especially in more complex problem instances. In this paper, we extend the histogram density estimation framework by introducing a model called clustgram, which uses arbitrary density functions as components of the density rather than just uniform components. The new model is based on finding a clustering of the sample points and determining the type of the density function for each cluster. We regard the problem of learning clustgrams as a model selection problem and use the theoretically appealing minimum description length principle for solving the task.

The development of medical domain applications has been one of the most active research areas recently. One example of a medical domain application is a detection system for heart disease based on computer-aided diagnosis methods, where the data is obtained from some other sources and is evaluated by computer based applications. Up to now, computers have usually been used to build knowledge based clinical decision support systems which used the knowledge from medical experts, and transferring this knowledge into computer algorithms was done manually. This process is time consuming and really depends on the medical expert’s opinion, which may be subjective. To handle this problem, machine learning techniques have been developed to gain knowledge automatically from examples or raw data. Here, a weighted fuzzy rule-based clinical decision support system (CDSS) is presented for the diagnosis of heart disease, automatically obtaining the knowledge from the patient’s clinical data. The proposed clinical decision support system for risk prediction of heart patients consists of two phases, (1) automated approach for generation of weighted fuzzy rules and decision tree rules, and, (2) developing a fuzzy rule-based decision support system. In the first phase, we have used the mining technique, attribute selection and attribute weightage method to obtain the weighted fuzzy rules. Then, the fuzzy system is constructed in accordance with the weighted fuzzy rules and chosen attributes. Finally, the experimentation is carried out on the proposed system using the datasets obtained from the UCI repository and the performance of the system is compared with the neural network-based system utilizing accuracy, sensitivity and specificity.