An adaptive algorithm for feature selection in pattern recognition

Abstract

With the most recent advances in bioinformatics, the amount of information available for analysing certain diseases has increased considerably. Specifically, the use of microarrays makes it possible to obtain information on genetic patterns. The analysis of this information requires the use of new computational models and the modification of existing models so that it becomes possible to work with such an elevated amount of data. This study will demonstrate the integration of an expression analysis in a case-based reasoning system that can apply data mining techniques to classify and obtain patterns that have been stored in a case database for leukaemia patients.