Laboratoire de Génie Informatique et d’Automatique de l’Artois

Mohamed amir ESSEGHIR

Ph.D. student
(Left the LGI2A in 2012)
Member of the research axes:
Mohamed amir ESSEGHIR -- Gilles GONCALVES -- Yahya SLIMANI
Adaptive Particle Swarm Optimizer for Feature Selection
11th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2010 , pp 226-233, Vol. 6283, Paisley , Scotland, UK , Springer, LNCS, September 1-3, septembre 2010
Mohamed amir ESSEGHIR
Effective Wrapper-Filter hybridization through GRASP Schemata
Journal of Machine Learning Research, JMLR, pp 45-54, Vol. 10, juin 2010
Mohamed amir ESSEGHIR -- Gilles GONCALVES -- Yahya SLIMANI
Memetic Feature Selection: Benchmarking Hybridization Schemata
5th International Conference on Hybrid Artificial Intelligence Systems, HAIS 2010, pp 351-358, Vol. Volume 6076, San Sebastián, Spain, SpringerLink, LNCS, June 23-25, juin 2010
Mohamed amir ESSEGHIR -- Yahya SLIMANI -- Gilles GONCALVES
Distributed Feature Selection: benchmarking collaboration protocol
6ème Colloque sur l'Optimisation et les Systèmes d'Information, COSI'2009, Annaba, Algérie , May 25-27, mai 2009
Mohamed amir ESSEGHIR -- Tienté HSU -- Gilles GONCALVES -- Yahya SLIMANI
A Cooperative feature selection approach based on island model
Colloque sur l'Optimisation et les Systèmes d'Information, COSI 2007, oran, algérie, juin 2007
Mohamed amir ESSEGHIR -- Tienté HSU -- Gilles GONCALVES -- Yahya SLIMANI
A feature selection approach based on parallel genetic algorithm for high dimensional data sets
Métaheuristiques 2006, META'06, Hammamet, Tunisie, novembre 2006

Author of the Ph.D. thesis "Metaheuristics for the feature selection problem : adaptive, memetic and swarm approaches"

2005 - 2011

Although the expansion of storage technologies, networking systems, and information system methodologies, the capabilities of conventional data processing techniques remain limited. The need to knowledge extraction, compact representation and data analysis are highly motivated by data expansion. Nevertheless, learning from data might be a complex task, particularly when it includes noisy, redundant and information-less attributes. Feature Selection (FS) tries to select the most relevant attributes from raw data, and hence guides the construction of final classification models or decision support systems. Selected features should be representative of the underlying data and provide effective usefulness to the targeted learning paradigm (i.e. classification). In this thesis, we investigate different optimization paradigms as well as its adaptation to the requirements of the feature selection challenges, namely the problem combinatorial nature. Both theoritical and empirical aspects were studied, and confirm the effectiveness of the adopted methodology as well as the proposed metaheuristic based approaches.

PlaiiMob (CISIT)

2007 - 2013

Plate-forme de simulation dédiée aux services de MOBilité

Summary :

The objective of this project is to illustrate how partial information can be exchanged in the context of inter-vehicle communication, for example to warn the driver of a potentially dangerous event (accident, obstacles on the road, braking, ...) or to assist him/her (find a free parking space, avoid traffic jams, be informed in real time of traffic conditions).