Service

GNU

JAVA

csv

eXiTCBR is a case-based reasoning (CBR) tool developed at the eXiT research group of the University of Girona. It goes beyond pure CBR prototyping and aims to support experimentation. The eXiT*CBR framework was designed to bring together CBR methods currently used in medical applications and data mining and visualization techniques that can be plugged into it. eXiT*CBR also facilitates the incorporation of new techniques, if required. eXiT*CBR architecture follows a modular approach based on the different phases in a CBR system. Each CBR steps is implemented as generic class. When a new method is required but not provided in the system, it can be assembled as a particular instance of a generic class. When other techniques need to be integrated or hybridized, the corresponding executable codes should also be included. It could be used to analize different configurations of a CBR system. The results achieved are obtained in form of plots that can be used for scientific purposes. For exploitation, a library could be generated that implements the choosed CBR system. A distributed CBR approach can be also designed. Plugins include genetic algorithms for feature relevance learning, and inheritance tools to interpred pedigree files.