Clustering to Enhance Case-Based Reasoning

In this article, we propose an approach to improve CBR processing mainly in its retrieval task. A major difficulty arise when founding several similar cases and consequently several solutions, hence a choice must be done involving an appropriate strategy focusing the best solution. This main difficulty has a direct impact on the adaptation task. To overcome this limitation many works related to the retrieval task were conducted as hybridizing CBR with data mining methods. Through this study, we provide a combining approach using CBR and clustering to reduce the search space in the retrieval step. The objective is to consider only the most interesting cases and the most interesting solution to support decision and provide an intelligent strategy that enables decision makers to have the best decision aid. We also present some preliminary results and suggestions to extend our approach.