A case retrieval approach using similarity and association knowledge

摘要: Retrieval is often considered the most important phase in Case-Based Reasoning (CBR), since it lays foundation for overall performance of CBR systems. aims to retrieve relevant cases that can be successfully used solving a new problem. To realize retrieval, systems typically rely on strategy exploits similarity knowledge, and called similarity-based retrieval (SBR). In SBR, knowledge approximates usefulness this paper, we show association analysis stored strengthen SBR. We present approach extracting representing from using rule mining. propose novel USIM-SCAR qualitatively enhances SBR by leveraging both knowledge. demonstrate significant advantages USIMSCAR over through an experimental evaluation medical datasets.