Research on Case Retrieval Model Based on Rough Set Theory and BP Neural Network

Retrieval is the key technology in case-based reasoning. It imposes a direct effect on the efficiency and quality of case-based reasoning, and the quality of the retrieved case determines the difficulty of case reuse and adaptation. In allusion to the traditional case retrieval technology disadvantage of die design, a case retrieval method based on rough set theory and neural work is presented. Firstly, The paper analyzes and deals with die case database using rough set theory, and it uses a method using grade classification and decision attributes support degree to deal with the quantitative features. And it confirms the important degree of all types of characteristic attributes. To aim to build up a retrieval method based on case´s key attributes. BP neural network is used to retrieve the similar case. The proposed method is also demonstrated by an application example. The technology guarantees the validity of case retrieval reduces system dependence and improves efficiency of case retrieval.