Research On Production Expert System And Reasoning Machine Based On Web Services

Expert system is an important branch of the artificial intelligence, which uses the reasoning machine to solve the difficulties and problems in the field through the professional knowledge in the brain of experts in the area. Its core lies in knowledge-base and reasoning machine.With deeply research on expert system, the following innovative results are given:1) As for maintenance of the knowledge-base, in order to guarantee the validity of knowledge-base and the consistency among the rule, and to improve the speed and correctness of reasoning, a successful maintenance method of knowledge-base is presented and the methods of checking the redundant rule, circulation rule, conflict rule and isolated rule are given.2) A new reasoning tree structure is provided, which, according to the process of reasoning in the forward reasoning machine, adopts hierarchy and classification in storing rule and greatly reduces the number of rule-matching per time when the machine reasons. So the reasoning efficiency is greatly improved. In this thesis, the updating algorithm for new reasoning tree structure (including insert and delete rule algorithm) and reasoning algorithm are demonstrated in detail.3) Web services technology and its advantages on Internet are introduced.With using component to encapsulate reasoning machine, maintenance of knowledge-base and other business logical functional modules,a distributed decision expert system frame based on web is designed,which have trans-platform ability by using web services' abilities of interconnection between different platform. In addition,a simple application related to students' department-choice is introduced to express its function,which has better extendibility, elasticity, compatibility and cooperation and also realized the cooperation of distributed multi expert systems.