Genetic Algorithm and Tabu Search for Feature Selection

Sabra EL FERCHICHI, Kaouther LAABIDI
LACS, ENIT Tunis
Tunisie

Salah ZIDI
LAGIS, USTL Lille
France

Abstract: In this paper we propose a wrapper approach to select features involving the Support Vector Machines (SVM) combined with a metaheuristic optimization algorithm: Tabu Search and Genetic Algorithms. Feature selection is efficient in searching the most descriptive features which would contribute in increasing the performance of the inductive algorithm by reducing dimensionality and processing time. The process we propose is based on the use of the rate of misclassification as an evaluating criterion. First, we used the tabu algorithm to guide the search of the optimal set of features; then a genetic algorithm is implemented to reach the same goal. This procedure is applied on data from regulation of urban transport network systems. A comparison between the performances of each search engine (TS and GAs) used is then presented.

Keywords: Features Selection, Support Vector Machines, Tabu Search, Genetic algorithm and urban transport regulation.

Sabra El Ferchichi received the engineering degree in Electrical Engineering and the M.S degree in Control and Signal Processing from the National School Engineering of Tunis in 2007 and 2008. Where, she is currently pursuing her Ph.D. degree in Department of Electrical Engineering at National School Engineering of Tunis. Her research interests include Feature extraction and selection, learning machines, metaheuristic algorithms.

Kaouther Laabidi was born in Tunis, Tunisia. She received the Master degree from the “Ecole Suprieur de Sciences et de Technologie de Tunis”, in 1995 and the Ph.D degree in Genie Electique from the “Ecole Nationale d’Ingnieurs de Tunis”, in 2005. She is currently preparing the ability degree in the laboratory ACS “Analyse et Commande des Systmes”.His research is related to the Ide ntification and control of complex systems.

Salah Zidi was born in Gafsa, Tunisia, in 1980. In 2003, he received the B.E. degree in electrical engineering from the National School of Engineers of Sfax, Tunisia. He received the D.E.A. degree in new technologies of dedicated computing systems from the National School of Engineers of Sfax, Tunisia, in 2004. He received the Ph.D. degrees in automatic control from the University of Lille, Lille, France, in 2007. He is currently a teacher and researcher at the University of Lille. Her research interests include transport, reconfiguration, scheduling, ant colony algorithms, and decision-support systems.

>Full text
CITE THIS PAPER AS:
Sabra EL FERCHICHI, Kaouther LAABIDI, Salah ZIDI, Genetic Algorithm and Tabu Search for Feature Selection, Studies in Informatics and Control, ISSN 1220-1766, vol. 18 (2), pp. 181-187, 2009.