Using dynamic Bayesian networks for the prediction of mental deficiency in children with down syndrome

摘要

This work is located in the domain of the Knowledge Discovery from Data (KDD). The purpose of the KDD is the extraction of knowledge or of a knowledge starting from great number of data which evolve in a dynamic way. In this work we propose an approach for the temporal KDD. The Bayesian Network (BN) is one of the techniques used in KDD. Our objective comes back to fix the best algorithm of incremental learning of structure extracted by the Dynamic Bayesian Network (DBN) and using it in the decision making in a dynamic way. Our scope of application is the case of Down Syndrome (DS) also known as trisomy 21, the data are provided by the medical genetics and Child Psychiatry units of the university hospital Hedi Chaker Sfax, Tunisia.