Band 23, Heft 3

Sutra is a form of text representation that has been used in both Tamil and Sanskrit literature to convey information in a short and crisp manner. Nanool , an ancient Tamil grammar masterpiece has used sutras for defining grammar rules. Similarly, in Sanskrit literature, many of the Shāstrās have used sutras for a concise representation of their content. Sutras are defined as short aphorisms, formulae-like structures that convey the complete essence of the text. They act as indices to the elaborate content they refer to. Inspired by their characteristics, this article proposes an indexing mechanism based on sutras for discourse structures built using rhetorical structure theory (RST) and also using Sangati , a concept proposed in Sanskrit literature. The indices identified by the indexer are ideal for question answering (QA), summary generation, and information retrieval (IR) systems. The indexer has been tested on IR system using 1000 Tamil language text documents. A performance comparison has also been made with one of the existing RST-based indexing technique.

A novel piecewise water flow technique for text line extraction from multi-skewed document images of handwritten text of different scripts is presented here. The basic water flow technique assumes that the hypothetical water flows from both left and right sides of the image frame. This flow of water fills up the gaps between consecutive objects (texts) but faces obstruction if any object lies in the path of the flow. All unwetted regions in the document image are then labeled distinctly to extract the text lines. However, the technique fails when two neighboring text lines touch each other, as water gets obstructed by the touching segment(s). To get rid of this difficulty, we have modified the basic water flow technique by iteratively applying the same over the vertically segmented document images. The main purpose of this vertical segmentation is to localize the text line segment(s) where two text lines get joined. These segments are then horizontally fragmented, and each fragment is placed suitably to the text line in which it actually belongs to. This way, the probable data loss during isolation of the touching text line segment is minimized. Both the techniques (current and basic ones) have been tested on three different databases, viz., CMATERdb 1.1.1, CMATERdb 1.1.2, and ICDAR2009 handwritten segmentation contest pages, respectively. The test results show that the present technique outperforms the basic one for all three databases.

Multiple sequence alignment (MSA) is one of the major problems that can be encountered in the bioinformatics field. MSA consists in aligning a set of biological sequences to extract the similarities between them. Unfortunately, this problem has been shown to be NP-hard. In this article, a new algorithm was proposed to deal with this problem; it is based on a quantum-inspired cuckoo search algorithm. The other feature of the proposed approach is the use of a randomized progressive alignment method based on a hybrid global/local pairwise algorithm to construct the initial population. The results obtained by this hybridization are very encouraging and show the feasibility and effectiveness of the proposed solution.

A large class of problems deals with temporal data. Identifying temporal patterns in these datasets is a natural as well as an important task. In recent times, researchers have reported an algorithm for finding calendar-based periodic pattern in time-stamped data without considering the purchased quantities of the items. However, most of the real-life databases are nonbinary, and therefore, exploring various calendar-based patterns (yearly, monthly, weekly, daily) with their purchased quantities may discover information useful to improve the quality of business decisions. In this article, a technique is proposed to extract calendar-based periodic patterns from nonbinary transactions. In this connection, the concept of certainty factor has been introduced by incorporating transaction frequency for overlapped intervals. Algorithms have been designed to mine frequent itemsets along with intervals and quantity. In addition to that, we have designed an algorithm to find the periodicity of the pattern. The algorithm is tested with real-life data, and the results are given.

Coreference resolution is a challenging natural language processing task, and it is difficult to identify the correct mentions of an entity that can be any noun or noun phrase. In this article, a semisupervised, two-stage pattern-based bootstrapping approach is proposed for the coreference resolution task. During Stage 1, the possible mentions are identified using word-based features, and during Stage 2, the correct mentions are identified by filtering the non-coreferents of an entity using statistical measures and graph-based features. Whereas the existing approaches use morphosyntactic and number/gender agreement features, the proposed approach uses semantic graph-based context-level semantics and nested noun phrases in the correct mentions identification. Moreover, mentions without the number/gender information are identified, using the context-based features of the semantic graph. The evaluation performed for the coreference resolution shows significant improvements, when compared with the word association-based bootstrapping systems.

Clustering plays an important role in data mining, pattern recognition, and machine learning. Then, single-valued neutrosophic sets (SVNSs) are a useful means to describe and handle indeterminate and inconsistent information, which fuzzy sets and intuitionistic fuzzy sets cannot describe and deal with. To cluster the data represented by single-value neutrosophic information, the article proposes a single-valued neutrosophic minimum spanning tree (SVNMST) clustering algorithm. Firstly, we defined a generalized distance measure between SVNSs. Then, we present an SVNMST clustering algorithm for clustering single-value neutrosophic data based on the generalized distance measure of SVNSs. Finally, two illustrative examples are given to demonstrate the application and effectiveness of the developed approach.

To assist in the evaluation process when determining architectures for new robots and intelligent systems equipped with artificial emotions, it is beneficial to understand the systems that have been built previously. Other surveys have classified these systems on the basis of their technological features. In this survey paper, we present a classification system based on a model similar to that used in psychology and philosophy for theories of emotion. This makes possible a connection to thousands of years of discourse on the topic of emotion. Five theories of emotion are described based on an emotion theory model proposed by Power and Dalgleish. The paper provides classifications using a model of 10 new questions, for 14 major research projects that describe implementations or designs for systems that use artificial emotions for either robotics or general artificial intelligence. We also analyze the trends in the usage of various theories and complexity changes over time.

An improved image enhancement approach via nonsubsampled contourlet transform (NSCT) is proposed in this article. We constructed a geometric image transform by combining nonsubsampled directional filter banks and a nonlinear mapping function. Here, the NSCT of the input image is first decomposed for L -levels and its noise standard deviation is estimated. It is followed by calculating the noise variance and threshold calculation, and computing the magnitude of the corresponding coefficients in all directional subbands. Then, the nonlinear mapping function is used to modify the NSCT coefficients for each directional subband, which keeps the coefficients of strong edges, amplifies the coefficients of weak edges, and zeros the noise coefficients. Finally, the enhanced image is reconstructed from the modified NSCT coefficients. Three experiments are carried out respectively on images from subjective vision quality and objective evaluation measures. The first experiment is the algorithm performed on images. The subsequent experiments are the information entropy and spatial frequency. The experimental results demonstrate that the proposed method can gain better performance in enhancing the low-contrast parts of an image while keeping its clear edges.