An Knowledge Acquisition Method Based on Rough Set in Plant Diagnostic Expert System

摘要

Knowledge acquisition is the bottleneck to develop expert system. It usually takes a long period to acquire plant disease knowledge using the traditional methods. Aiming at this problem, this paper describes relations between rough set theory and rule-based description of plant diseases, which corresponds to the process of knowledge acquisition of expert system. Then the exclusive rules, inclusive rules and disease images of rapeseed disease are built based on the PDES diagnosis model, and the definition of probability rule is put forward. At last, the paper presents the rule-based automated induction reasoning method, including exhaustive search, post-processing procedure, estimation for statistic test and the bootstrap and resampling methods. We also introduce automated induction of the rule-based description, which is used in our plant diseases diagnostic expert system. The experimental results not only show that rough set theory gives a very suitable framework to represent processes of uncertain knowledge extraction, but also that this method induces diagnostic rules correctly. This method can act as the assistant tool for development of diagnosis expert system, and has a extensive application in intelligent information systems.