Skip to Main content Skip to Navigation

Un Système de Classification Supervisée à Base de Règles Implicatives

Abstract : This PhD thesis presents a series of research works done in the field of supervised data classification, more precisely in the domain of semi–automatic learning of fuzzy rules–based classifiers. The prepared manuscript presents first an overview of the classification problem, and also of the main classification methods that have already been implemented and certified in order to place the proposed method in the general context of the domain. Once the context established, the actual research work is presented : the definition of a formal background for representing an elementary fuzzy rule-based classifier in a bidimensional space, the description of a learning algorithm for these elementary classifiers for a given data set and the conception of a multi-dimensional classification system which is able to handle multi–classes problems by combining the elementary classifiers. The implementation and testing of all these functionalities and finally the application of the resulted classifier on two real–world digital image problems are finaly presented : the analysis of the quality of industrial products using 3D tomographic images and the identification of regions of interest in radar satellite images.
Complete list of metadata

Cited literature [159 references]  Display  Hide  Download
Contributor : Sylvie Galichet Connect in order to contact the contributor
Submitted on : Friday, October 30, 2015 - 2:21:32 PM
Last modification on : Friday, November 6, 2020 - 3:26:58 AM
Long-term archiving on: : Friday, April 28, 2017 - 5:36:43 AM


  • HAL Id : tel-01222735, version 1



Lavinia Darlea. Un Système de Classification Supervisée à Base de Règles Implicatives. Intelligence artificielle [cs.AI]. Université de Savoie, 2010. Français. ⟨tel-01222735⟩



Record views


Files downloads