Knowledge based supervised classification: an application to image processing

6XPPDU\ We present a knowledge based supervised classification method. Our modelisation is based on automatic generation of classification rules. The classification function is directly given in the form of production rules base. The proposed learning method is multi-features, it allows to take into account the possible predictive power of a simultaneously considered features conjunction. On the other hand, the feature space partition allows a multi-valued representation of the features and data imprecision integration. The rules conclusions are accompanied by belief degrees. This uncertainty is managed in the learning phase as well as in the recognition one. To introduce more flexibility and overcome the boundary problem due to the discretisation, we propose to use approximate reasoning. We introduce, in this purpose, an adequate distance to compare neighboring facts. This distance, measuring imprecision, combined with uncertainty of classification decisions represented by belief degrees, drives the approximate inference. The proposed method was implemented in a tool called SUCRAGE and confronted with a real application in the field of image processing. The obtained results are very satisfactory. They validate our approach and allow us to consider other application fields. Facing the increase of data amount recorded daily, the detection of both structures and specific links between them, the organisation and the search of exploitable knowledge in this information become a strategic stake for decision holding and prediction task. This complex problem, also known as « Data Mining » has multiple aspects. We focus on one of them : supervised learning. We propose a learning method from examples situated at the junction of statistical methods and those based on Artificial Intelligence techniques. Our modelisation is based on automatic generation of classification rules. The classification function is directly given in the form of production rules base. This ensures the transparency and easy interpretation of the classifier. The construction of production rules ,)) >SUHPLVH@ 7+(11 >FRQFOXVLRQ@ using the knowledge and the know-how of an expert is a very difficult task. The complexity and cost of such a knowledge acquisition have led to an important development of learning methods used for an automatic knowledge extraction [9] [7]. In the pattern recognition domain, expert's rules allow to determine the belonging of a pixel to a class. For instance, in a human thigh cryosection image a pixel will be classified as bone, muscle, ... In the medical domain, it is practically impossible to obtain from an image classification given by …