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

Ph.D. thesis of Mohamed Amir ESSEGHIR

Metaheuristics for the feature selection problem : adaptive, memetic and swarm approaches

Starting date: 1 October 2005
Keywords: Feature selection, Metaheuristics, Classification, Combinatorial optimization, Genetic algorithms, GRASP, PSO, Local search
Advising:

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.

Involved research themes:

Involved application areas:

No partner is associated with this element.

Defense

Defense took place the 29/11/2011 am30 10:00 Prestige room - FSA - Béthune

Jury:

  • Rapporteur Patrick SIARRY Université Paris-Est Créteil
  • Rapporteur Jean-Charles CREPUT Université de Technologie de Belfort-Montbéliard
  • Examinateur Rémy DUPAS Université Bordeaux I
  • Examinateur Laetitia JOURDAN INRIA / USTL
  • Examinateur Daniel JOLLY Université d'Artois