Case retrieval is a complex and important stage of the overall CBR process, involving situation assessment and a flexible combination of case memory search and matching. The goal of the paper is to discuss a retrieval approach in a case library organized through classes of prototypes. Situation assessment is realized with a Bayesian classification step, aimed at defining a uniform framework for feature evaluation and at restricting search only to relevant parts of the case library. Classical K-NN methods or pruning based technique like Pivoting-Based Retrieval can then be applied to retrieve and match cases, by considering intra-class or inter-class retrieval. The proposed method has been evaluated within a decision support system for the therapy revision of patients affected by Diabetes Mellitus.
Retrieval in a prototype-based case library: A case study in diabetes therapy revision
BELLAZZI, RICCARDO;
1998-01-01
Abstract
Case retrieval is a complex and important stage of the overall CBR process, involving situation assessment and a flexible combination of case memory search and matching. The goal of the paper is to discuss a retrieval approach in a case library organized through classes of prototypes. Situation assessment is realized with a Bayesian classification step, aimed at defining a uniform framework for feature evaluation and at restricting search only to relevant parts of the case library. Classical K-NN methods or pruning based technique like Pivoting-Based Retrieval can then be applied to retrieve and match cases, by considering intra-class or inter-class retrieval. The proposed method has been evaluated within a decision support system for the therapy revision of patients affected by Diabetes Mellitus.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.