Mohand Tahar Kechadi

Customer churn prediction in telecommunications

This paper presents a new set of features for land-line customer churn prediction, including 2 six-month Henley segmentation, precise 4-month call details, line information, bill and payment information, account information, demographic profiles, service orders, complain information, etc. Then the seven prediction techniques (Logistic Regressions, Linear Classifications, Naive Bayes, Decision Trees, Multilayer Perceptron Neural Networks, Support Vector Machines and the Evolutionary Data Mining Algorithm) are applied in customer churn as predictors, based on the new features. Finally, the comparative experiments were carried out to evaluate the new feature set and the seven modelling techniques for customer churn prediction. The experimental results show that the new features with the six modelling techniques are more effective than the existing ones for customer churn prediction in the telecommunication service field.

Preprocessing Techniques for Online Handwriting Recognition

This paper proposes a new preprocessing technique for online handwriting. The approach is to first remove the hooks of the strokes by using changed-angle threshold with length threshold, then filter the noise by using a smoothing technique, which is the combination of the cubic spline and the equal-interpolation methods. Finally, the handwriting is normalised. Experiments are carried out using the benchmark UNIPEN database. The experimental results show that our preprocessing technique can improve the recognition rates by at least 10%.

Towards a New Data Mining-Based Approach for Anti-Money Laundering in an International Investment Bank

Today, money laundering (ML) poses a serious threat not only to financial institutions but also to the nation. This criminal activity is becoming more and more sophisticated and seems to have moved from the cliche of drug trafficking to financing terrorism and surely not forgetting personal gain. Most international financial institutions have been implementing anti-money laundering solutions (AML) to fight investment fraud. However, traditional investigative techniques consume numerous man-hours. Recently, data mining approaches have been developed and are considered as well-suited techniques for detecting ML activities. Within the scope of a collaboration project for the purpose of developing a new solution for the AML Units in an international investment bank based in Ireland, we propose a new data mining-based approach for AML. In this paper, we present this approach and some preliminary results associated with this method when applied to transaction datasets.

Online Acquisition of Digital Forensic Evidence

Providing the ability to any law enforcement officer to remotely transfer an image from any suspect computer directly to a forensic laboratory for analysis, can only help to greatly reduce the time wasted by forensic investigators in conducting on-site collection of computer equipment. RAFT (Remote Acquisition Forensic Tool) is a system designed to facilitate forensic investigators by remotely gathering digital evidence. This is achieved through the implementation of a secure, verifiable client/server imaging architecture. The RAFT system is designed to be relatively easy to use, requiring minimal technical knowledge on behalf of the user. One of the key focuses of RAFT is to ensure that the evidence it gathers remotely is court admissible. This is achieved by ensuring that the image taken using RAFT is verified to be identical to the original evidence on a suspect computer.

A Fast Feature Selection Model for Online Handwriting Symbol Recognition

Many feature selection models have been proposed for online handwriting recognition. However, most of them require expensive computational overhead, or inaccurately find an improper feature set which leads to unacceptable recognition rates. This paper presents a new efficient feature selection model for handwriting symbol recognition by using an improved sequential floating search method coupled with a hybrid classifier, which is obtained by combining hidden Markov models with multilayer forward network. The effectiveness of proposed method is verified by comprehensive experiments based on UNIPEN database

Automatic multi-objective clustering based on game theory

Novel multi-objective clustering algorithm based-sequential games was proposed.It optimizes simultaneously and efficiently multiple conflicting objectives.Proposed approach can automatically calculate the optimal number of clusters.This algorithm shows good performance of generating high-quality solutions.Experimental study demonstrates the effectiveness of our algorithm over others. Data clustering is a very well studied problem in machine learning, data mining, and related disciplines. Most of the existing clustering methods have focused on optimizing a single clustering objective. Often, several recent disciplines such as robot team deployment, ad hoc networks, multi-agent systems, facility location, etc., need to consider multiple criteria, often conflicting, during clustering. Motivated by this, in this paper, we propose a sequential game theoretic approach for multi-objective clustering, called ClusSMOG-II. It is specially designed to optimize simultaneously intrinsically conflicting objectives, which are inter-cluster/intra-cluster inertia and connectivity. This technique has an advantage of keeping the number of clusters dynamic. The approach consists of three main steps. The first step sets initial clusters with their representatives, whereas the second step calculates the correct number of clusters by resolving a sequence of multi-objective multi-act sequential two-player games for conflict-clusters. Finally, the third step constructs homogenous clusters by resolving sequential two-player game between each cluster representative and the representative of its nearest neighbor. For each game, we define payoff functions that correspond to the model objectives. We use a methodology based on backward induction to calculate a pure Nash equilibrium for each game. Experimental results confirm the effectiveness of the proposed approach over state-of-the-art clustering algorithms.

Forensic Acquisition of IMVU: A Case Study

There are many applications available for personal computers and mobile devices that facilitate users in meeting potential partners. There is, however, a risk associated with the level of anonymity on using instant message applications, because there exists the potential for predators to attract and lure vulnerable users. Today Instant Messaging within a Virtual Universe (IMVU) combines custom avatars, chat or instant message (IM), community, content creation, commerce, and anonymity. IMVU is also being exploited by criminals to commit a wide variety of offenses. However, there are very few researches on digital forensic acquisition of IMVU applications. In this paper, we discuss first of all on challenges of IMVU forensics. We present a forensic acquisition of an IMVU 3D application as a case study. We also describe and analyse our experiments with this application.

Clustering Based on Sequential Multi-Objective Games

We propose a novel approach for data clustering based on sequential multi-objective multi-act games (ClusSMOG). It automatically determines the number of clusters and optimises simultaneously the inertia and the connectivity objectives. The approach consists of three structured steps. The first step identifies initial clusters and calculates a set of conflict-clusters. In the second step, for each conflict-cluster, we construct a sequence of multi-objective multi-act sequential two-player games. In the third step, we develop a sequential two-player game between each cluster representative and its nearest neighbour. For each game, payoff functions corresponding to the objectives were defined. We use a backward induction method to calculate Nash equilibrium for each game. Experimental results confirm the effectiveness of the proposed approach over state-of-the-art clustering algorithms.

Classical Mechanics Optimization for Image Segmentation

In this work, we focus on image segmentation by simulating the natural phenomenon of the bodies moving through space. For this, a subset of image pixels is regularly selected as planets and the rest as satellites. The attraction force is defined by Newton’s third law (gravitational interaction) according to the distance and color similarity. In the first phase of the algorithm, we seek an equilibrium state of the earth-moon system in order to achieve the second phase, in which we search an equilibrium state of the earth-apple system. As a result of these two phases, bodies in space are constructed; they represent segments in the image. The objective of this simulation is to find and then extract the multiple segments from an image.

Green Approach for Joint Management of Geo-Distributed Data Centers and Interconnection Networks

Every time an Internet user downloads a video, shares a picture, or sends an email, his/her device addresses a data center and often several of them. These complex systems feed the web and all Internet applications with their computing power and information storage, but they are very energy hungry. The energy consumed by Information and Communication Technology (ICT) infrastructures is currently more than 4% of the worldwide consumption and it is expected to double in the next few years. Data centers and communication networks are responsible for a large portion of the ICT energy consumption and this has stimulated in the last years a research effort to reduce or mitigate their environmental impact. Most of the approaches proposed tackle the problem by separately optimizing the power consumption of the servers in data centers and of the network. However, the Cloud computing infrastructure of most providers, which includes traditional telcos that are extending their offer, is rapidly evolving toward geographically distributed data centers strongly integrated with the network interconnecting them. Distributed data centers do not only bring services closer to users with better quality, but also provide opportunities to improve energy efficiency exploiting the variation of prices in different time zones, the locally generated green energy, and the storage systems that are becoming popular in energy networks. In this paper, we propose an energy aware joint management framework for geo-distributed data centers and their interconnection network. The model is based on virtual machine migration and formulated using mixed integer linear programming. It can be solved using state-of-the art solvers such as CPLEX in reasonable time. The proposed approach covers various aspects of Cloud computing systems. Alongside, it jointly manages the use of green and brown energies using energy storage technologies. The obtained results show that significant energy cost savings can be achieved compared to a baseline strategy, in which data centers do not collaborate to reduce energy and do not use the power coming from renewable resources.