Volume 20, Issue 4

In this paper, we propose a person verification scheme using a novel combination of palmprint and speech. The crucial aspect of biometric based verification lies in its use of features in verification. Thus, in this paper, we propose two novel feature extraction methods for palmprint verification. The proposed methods are based on the Gaussian mixture model followed by subspace based approaches such as ICA I and ICA II, called independent component analysis I mixture model (ICA I MM) and independent component analysis II mixture model (ICA II MM) for palmprint verification. The speech verification system uses sphericity based measurement for person verification. The proposed multimodal biometric system uses a fairly large database of 150 users. The robustness of the proposed methods is validated by introducing six different types of noises on palmprint images and Gaussian white noise with two different signal-to-noise ratios on speech samples. Then, fusion of palmprint and speech are carried out at match score level using the weighted SUM rule for both clean and noisy conditions. Finally, we compare the results of combined biometric with the results of individual biometric and also the results of the proposed methods against conventional (without mixture model) subspace methods.

Due to existence of a huge amount of textual data either on the World Wide Web or in textual databases like PubMed, the development of novel automatic keyphrase extraction methods has emerged as one of the key research problems in recent past. Consequently, a number of machine learning techniques, mostly supervised, have been proposed to extract keyphrases from text documents. But, one of the main bottlenecks that hinders the success of such systems is the requirement of annotated corpora for training purpose. In this paper, we propose the design of a deep text mining system to identify keyphrases in text documents that are either unstructured or semi-structured in nature. The novelty of our system lies in its applicability on a single document, instead of demanding a collection of annotated texts for training, to identify keyphrases embedded within it. The proposed system applies parsing techniques to identify candidate phrases. After mapping the original set of candidate phrases into a low-dimensional space using Singular Value Decomposition (SVD), the Markov Clustering (MCL) technique is applied to cluster related sentences together. Finally, considering each cluster as a document, Latent Dirichlet Allocation (LDA) is applied to identify feasible keyphrases that are presented to users in non-increasing order of their relevance score values. The efficacy of the proposed system is established through experimentation on datasets from two different domains. On comparative evaluation, we found that the proposed system outperforms KEA and KEA that apply the supervised machine learning approach for automatic keyphrase extraction from text documents.

Biometrics is essentially a pattern recognition system that recognizes an individual using their unique anatomical or behavioral patterns such as face, fingerprint, iris, signature etc. Recent researches have shown that many biometric traits are vulnerable to spoof attacks. Moreover, recent works showed that, contrary to a common belief, multimodal biometric systems in parallel fusion mode can be intruded even if only one trait is spoofed. However, most of the results were obtained using simulated spoof attacks, under the assumption that the spoofed and genuine samples are indistinguishable, which may not be true for all biometric traits. In addition, so far vulnerability of multimodal biometric systems in serial fusion mode against spoof attacks has not been investigated. These issues raise a demand to investigate the robustness of multimodal systems under realistic spoof attacks. In this paper, we empirically investigate the performance of serial and parallel biometric fusion modes under realistic spoof attacks. Preliminary empirical results on real biometric systems made up of face, fingerprint and iris confirm that multimodal biometric systems in both fusion modes are not intrinsically robust against spoof attacks as believed so far. In particular, multimodal biometric systems in serial fusion mode can be even less robust than systems in parallel mode. We also experimentally found that incorporating the biometric sample quality in biometric fusion increases the robustness of the multimodal systems against spoof attacks. In the end, we study the trade-off between performance and robustness of the biometric systems under spoof attacks.

A number of conditioning experiments utilize food as a reward. Hunger is considered to be a critical factor governing the animal's behavior in these experiments. Despite its significance, most theories of animal conditioning fail to take hunger into consideration while analyzing the behavioral data. In this paper, we analyze the neuroscientific data supporting the hypothesis that hunger and food consumption affect the brain's dopamine system, which in turn governs the animal's behavior. According to this hypothesis, chronic hunger results in a decrease in the extra-cellular dopamine levels in the animal's brain. This decrease is believed to trigger a series of reactions that increase the responsivity of the phasic dopamine system. A direct consequence of this is an increased vigor of all dopamine-dependent behaviors. The level of extra-cellular dopamine is also modulated by the process of food consumption via the neurotransmitter Acetylcholine. Food consumption raises the dopamine above the baseline level. This rise depends on the animals' hunger, which when satisfied increases the level of Acetylcholine, which causes the dopamine level to fall back to the baseline. Thus, extra-cellular dopamine governs the response vigor, with an increase in dopamine resulting in a more vigorous response. This paper makes two primary contributions. Firstly, we present an abstract mathematical model based on the above hypothesis. Our mathematical model is able to provide a simple explanation for a number of behavioral findings. Another contribution of this paper is the development of a neurocomputational Leabra model of dopamine and acetylcholine activity in the basal ganglia to incorporate hunger and satiation. The experimental results obtained from this neural model are also largely in agreement with behavioral findings.

In multiple robot systems, the problem of allocation of complex tasks to heterogeneous teams of robots, also known as the multiple robot coalition formation problem, has begun to receive considerable attention. Efforts to address the problem range from heuristics based approaches that search the subspaces of the coalition structure to evolutionary learning approaches. Conventional approaches typically strive to optimize a single objective function such as the number of tasks executed or the time required to execute all tasks, or a weighted function of such objectives. In real world applications, objectives such as minimizing distance traveled and maximizing the number of tasks completed are often conflicting in nature. The coalition formation problem thus naturally lends itself to a multi-objective optimization approach based on evolutionary learning. In this paper, we formulate the problem of mapping coalitions of robots to a set of tasks as a multi-objective optimization problem and propose a variant of non-dominated sorting genetic algorithm (NSGA-II) to arrive at trade-off solutions. Additionally, we extend the solution to domains where robot resources are non-additive. Simulations demonstrate the efficacy of the proposed approach in generating the set of Pareto-optimal solutions.