Networks have provided a representation for a wide range of real systems, including communication networks, money transfer networks and biological systems. Communities repre-sent fundamental structures for understanding the organization of real-world networks. Uncovering coherent groups in these networks is the goal of community detection. A community is a mesoscopic structure with nodes heavily connected in their groups by comparison to the nodes in other groups. Commu-nities might also overlap as they may share one or multiple nodes. This paper lays the foundation for an application on transactional multigraphs (networks of financial transactions in which nodes can be linked with multiple edges), through the discovery of communities. Due to their complexity, our goal is to find the most effective way of simplifying multigraphs to weighted graphs, while preserving properties of the network. We tested five weights' calculation function and community detection algorithms were applied. A comparison of the outputs based on extrinsic and intrinsic evaluation metrics is then held.

Networks have provided a representation for a wide range of real systems, including communication flow, money transfer or biological systems, to mention just a few. Communities represent fundamental structures for understanding the organization of real-world networks. Uncovering coherent groups in these networks is the goal of community detection. A community is a mesoscopic structure with nodes heavily connected within their groups by comparison to the nodes in other groups. Communities might also overlap as they may share one or multiple nodes. Evaluating the results of a community detection algorithm is an equally important task. This paper introduces metrics for evaluating overlapping community detection. The idea of introducing new metrics comes from the lack of efficiency and adequacy of state-of-the-art metrics for overlapping communities. The new metrics are tested both on simulated data and standard datasets and are compared with existing metrics.

The electronic payment transaction involves the use of a smart card. A card application is a software, corresponding to standards and non-proprietary and proprietary specifications, and is stored in the smart card. Despite increased security with Euro pay Mastercard Visa (EMV) specifications, attacks still exist due to anomalies in the card application. The validation of the card application enables the detection of any anomaly, improving the overall security of electronic payment transactions. Among the different ways of validating a card application, we can use the verification of required behaviors. These behavior can be materialized as properties of commands sent by the terminal and responses from the smart card, using the Application Protocol Data Unit (APDU) from the ISO/IEC 7816 standard [1]. However, the creation of these behaviors is complicated. We propose in this article a way to automatically create such behaviors by using a genetic algorithm technique.

Mobile Money Transfer (MMT) services provided by mobile network operators enable funds transfers made on mobile devices of end-users, using digital equivalent of cash (electronic money) without any bank accounts involved. MMT simplifies banking relationships and facilitates financial inclusion, and, therefore, is rapidly expanding all around the world, especially in developing countries. MMT systems are subject to the same controls as those required for financial institutions, including the detection of Money Laundering (ML) - a source of concern for MMT service providers. In this paper we focus on an often practiced ML technique known as micro-structuring of funds or smurfing and introduce a new method for detection of fraud chains in MMT systems. Whereas classical detection methods are based on machine learning and data mining, this work builds on Predictive Security Analysis at Runtime (PSA@R), a model-based approach for event-driven process analysis. We provide an extension to PSA@R which allows us to identify fraudsters in an MMT service monitoring network behavior of its end-users. We evaluate our method on simulated transaction logs, containing approximately 460,000 transactions for 10,000 end-users, and compare it with classical fraud detection approaches. With 99.81% precision and 90.18% recall, we achieve better recognition performance in comparison with the state of the art.

We propose in this paper a low cost multimodal biometric system combining keystroke dynamics and 2D face recognition. The objective of the proposed system is to be used while keeping in mind: good performances, acceptability, and espect of privacy. Different fusion methods have been used (min, max, mul, svm, weighted sum configured with genetic algorithms, and, genetic programming) on the scores of three keystroke dynamics algorithms and two 2D face recognition ones. This multimodal biometric system improves the recognition rate in comparison with each individual method. On a chimeric database composed of 100 individuals, the best keystroke dynamics method obtains an EER of 8.77%, the best face recognition one has an EER of 6.38%, while the best proposed fusion system provides an EER of 2.22%.

We propose in this paper to study different color spaces for representing an image for the face authentication application. We used a generic algorithm based on a matching of keypoints using sift descriptors computed on one color component. Ten color spaces have been studied on four large and significant benchmark databases (ENSIB, FACES94, AR and FERET). We show that all color spaces do not provide the same efficiency and the use of the color information allows an interesting improvement of verification results.

Many object descriptors have been proposed in the state of the art. For many reasons (occlusion, point of view, acquisition conditions...), local descriptors have a better robustness for image understanding applications. The goal of this paper is to make a comparative study of eight recent local descriptors. The objective is here to quantify their ability to generate automatically an object taxonomy. In order to answer this question, we use the Caltech256 benchmark which provides a large object taxonomy used as reference. This study shows that SIFT, differential invariants and shape context descriptors are the best ones to achieve this goal.

International audience

This paper analyzes the benefit and the limitations of using a particular biometric technology "namely face authentication" for banking applications. We present first the general concepts of banking. We propose a method in order to replace the PIN code authentication by using biometrics data. Biometric authentication is then detailed. A face recognition method we developed is presented revealing as itself as a biometric candidate solution. We show the benefit and limits of this approach to be used in a real industrial context.