Research And Application Of Uncertainty Reasoning In Expert System

The emergence of artificial intelligence is a significant contribution to the history of science, and the emergence of expert system is an important milestone of artificial intelligence theory into reality. The emergence of expert systems has a major impact on production and human life.in modern times, it can be said that as long as there is demand, the expert system can work in any human work areas. Because of this, the in-depth study of the expert system has a broad prospect and practical significance. Domestic and international current expert system has been involved in many areas, such as medical areas, geological exploration, agriculture. Food problem is the main issue to world's population,, therefore, a large number of agricultural research and application of expert system on the economic and social development made a significant contribution to the society. Currently, expert system research is focused on knowledge representation and acquisition, architecture and inference mechanism. This article will focus on reasoning mechanisms of expert system t.Uncertainty reasoning is an very important part of an expert system,At present, based on different theories of uncertain reasoning models have been widely used among the various expert systems.This paper describes two typical expert system model of uncertain reasoning in detail:uncertainty reasoning based on probability theory and uncertainty reasoning based on credibility, and then introduce a detailed description of set pair analysis theory based on the uncertainty reasoning model. the difference between The reasoning and based on other theories of uncertainty reasoning is, based on set pair analysis of uncertain reasoning is not blindly to emphasize the uncertainty into certainty processing, but by contact dialectical perspective, regard the certainty and uncertainty as a system, give an integrated approach to deal with uncertainty information. Therefore it is better to deal with the real world, fuzzy and so on uncertain information, thus it has a unique advantage of dealing with the uncertainty of the real world:to enhance the reliability of the conclusions of reasoning machine launch and launch Conclusion speed increase. In this paper,after analysis and uncertainty reasoning of based on set pair analysis, we proposed idea of improvement, that is we make bold assumptions on the range factor j, thereby to speed up reasoning speed while ensuring reasoning reliability.finally, a detailed set of experimental results show that the analysis based on uncertainty reasoning is of feasibility in practical application and can speed up the reasoning.