Using clustering for maintaining case based reasoning systems

The success of the Case Based Reasoning system depends on the quality of case data and the speed of the retrieval process that can be expensive in time especially when the number of cases gets large. To guarantee this quality, maintaining the contents of a case base becomes necessary. This paper presents two case base maintenance methods. They are mainly based on the idea that the clustering analysis to a large case base can efficiently build new case bases, which are smaller in size and can easily use simpler maintenance operations. One of method is based on partitioning clustering technique and the other one on density clustering technique. Experiments are provided to show the effectiveness of our methods taking into account the performance criteria of the case base. In addition, we support our empirical evaluation with using a new criterion called “competence” in order to show the efficiency of our methods in building high-quality case bases while preserving the competence of the case bases.