Case Based Reasoning (CBR) Methodology for Car Fault Diagnosis System (CFDS) Using Decision Tree and Jaccard Similarity Method

摘要

Case Based Reasoning (CBR) is a general AI technique used for problem-solution, fault detection and diagnosis, reasoning, learning and decision support. It finds numerous applications in almost all service domains including industry, consumer electronics, automation, help desks and medical diagnosis. This paper proposes a system for car fault diagnosis (CFD). CFDS is designed based on case based reasoning (CBR) methodology and it uses Decision tree and Jaccard Similarity Method. Decision tree is used to store cases into Case Base (CB); and Jaccard Similarity Method is used to calculate similarity between new case and stored cases. CBR methodology helps the CFDS to retrieve the most similar cases from previously stored cases in CB. This paper focuses on clustering of cases into decision tree, stored in the CB and responding user with the solutions according to user's query case. To improve the functionality of the system, the conventional CBR cycle of Aamodt and Plaza has been modified.