- Journal Details
- First Published
- 30 Dec 2014
- Publication timeframe
- 4 times per year
- Open Access
Pulse Shape Discrimination of Neutrons and Gamma Rays Using Kohonen Artificial Neural Networks
Page range: 77 - 88
The potential of two Kohonen artificial neural networks I ANNs) - linear vector quantisa - tion (LVQ) and the self organising map (SOM) - is explored for pulse shape discrimination (PSD), i.e. for distinguishing between neutrons (n's) and gamma rays (γ’s). The effect that la) the energy level, and lb) the relative- of the training and lest sets, have on iden- tification accuracy is also evaluated on the given PSD datasel The two Kohonen ANNs demonstrate compfcmentary discrimination ability on the training and test sets: while the LVQ is consistently mote accurate on classifying the training set. the SOM exhibits higher n/γ identification rales when classifying new paltms regardless of the proportion of training and test set patterns at the different energy levels: the average tint: for decision making equals ∼100 /e in the cax of the LVQ and ∼450 μs in the case of the SOM.
- Open Access
On the Influence of Topological Characteristics on Robustness of Complex Networks
Page range: 89 - 100
In this paper, we explore the relationship between the topological characteristics of a complex network and its robustness to sustained targeted attacks. Using synthesised scale-free, small-world and random networks, we look at a number of network measures, including assortativity, modularity, average path length, clustering coefficient, rich club profiles and scale-free exponent (where applicable) of a network, and how each of these influence the robustness of a network under targeted attacks. We use an established robustness coefficient to measure topological robustness, and consider sustained targeted attacks by order of node degree. With respect to scale-free networks, we show that assortativity, modularity and average path length have a positive correlation with network robustness, whereas clustering coefficient has a negative correlation. We did not find any correlation between scale-free exponent and robustness, or rich-club profiles and robustness. The robustness of small-world networks on the other hand, show substantial positive correlations with assortativity, modularity, clustering coefficient and average path length. In comparison, the robustness of Erdos-Renyi random networks did not have any significant correlation with any of the network properties considered. A significant observation is that high clustering decreases topological robustness in scale-free networks, yet it increases topological robustness in small-world networks. Our results highlight the importance of topological characteristics in influencing network robustness, and illustrate design strategies network designers can use to increase the robustness of scale-free and small-world networks under sustained targeted attacks.
- Open Access
Integrated Statistical and Rule-Mining Techniques for Dna Methylation and Gene Expression Data Analysis
Page range: 101 - 115
For determination of the relationships among significant gene markers, statistical analysis and association rule mining are considered as very useful protocols. The first protocol identifies the significant differentially expressed/methylated gene markers, whereas the second one produces the interesting relationships among them across different types of samples or conditions. In this article, statistical tests and association rule mining based approaches have been used on gene expression and DNA methylation datasets for the prediction of different classes of samples (viz., Uterine Leiomyoma/class-formersmoker and uterine myometrium/class-neversmoker). A novel rule-based classifier is proposed for this purpose. Depending on sixteen different rule-interestingness measures, we have utilized a Genetic Algorithm based rank aggregation technique on the association rules which are generated from the training set of data by Apriori association rule mining algorithm. After determining the ranks of the rules, we have conducted a majority voting technique on each test point to estimate its class-label through weighted-sum method. We have run this classifier on the combined dataset using 4-fold cross-validations, and thereafter a comparative performance analysis has been made with other popular rulebased classifiers. Finally, the status of some important gene markers has been identified through the frequency analysis in the evolved rules for the two class-labels individually to formulate the interesting associations among them.
- Open Access
A Big-Bang Big-Crunch Optimized General Type-2 Fuzzy Logic Approach for Multi-Criteria Group Decision Making
Page range: 117 - 132
Multi-Criteria Group Decision Making (MCGDM) aims to find a unique agreement from a number of decision makers/users by evaluating the uncertainty in judgments. In this paper, we present a General Type-2 Fuzzy Logic based approach for MCGDM (GFLMCGDM). The proposed system aims to handle the high levels of uncertainties which exist due to the varying Decision Makers’ (DMs) judgments and the vagueness of the appraisal. In order to find the optimal parameters of the general type-2 fuzzy sets, we employed the Big Bang-Big Crunch (BB-BC) optimization. The aggregation operation in the proposed method aggregates the various DMs opinions which allow handling the disagreements of DMs’ opinions into a unique approval. We present results from an application for the selection of reading lighting level in an intelligent environment. We carried out various experiments in the intelligent apartment (iSpace) located at the University of Essex. We found that the proposed GFL-MCGDM effectively handle the uncertainties between the various decision makers which resulted in producing outputs which better agreed with the users’ decision compared to type 1 and interval type 2 fuzzy based systems.
- Open Access
Segmentation and Edge Detection Based on Modified ant Colony Optimization for Iris Image Processing
Page range: 133 - 141
Ant colony optimization (stocktickerACO) is a meta-heuristic algorithm inspired by food searching behavior of real ants. Recently stocktickerACO has been widely used in digital image processing. When artificial ants move in a discrete habitat like an image, they deposit pheromone in their prior position. Simultaneously, vaporizing of pheromone in each iteration step avoids from falling in the local minima trap. Iris recognition because of its great dependability and non-invasion has various applications. simulation results demonstrate stocktickerACO algorithm can effectively extract the iris texture. Also it is not sensitive to nuisance factors. Moreover, stocktickerACO in this research preserves details of the various synthetic and real images. Performance of ACO in iris segmentation is compared with operation of traditional approaches such as canny, robert, and sobel edge detections. Experimental results reveal high quality and quite promising of stocktickerACO to segment images with irregular and complex structures.