Improvement To Semantic Nets About Ambiguity And Uncertain Inference Based On Petri Nets

Semantic Nets (SN) has powerful capability of Knowledge Representation (KR), but there is no accepted Formal Description Architecture of it38- 39. This problem leads to different meanings of it and uncertainty of inference process. This paper improves SN inference based on Petri Nets (PN).It is needed to transform SN to PN for reasoning with PN: SN is transformed to Formal Logic(FL) in the first step. And then, FL is transformed to PN.Based on the principle of Resolution Refutation in FL, the inferring process of First Order Predication Logic (FOPL) is performed and SN Inference based on PN is implemented.Inference methods of several kinds of special forms of FOPL and searching strategies of Inference Algorithm is also discussed.Furthermore, a case study of the Courses Tutorial Expert System (CTES) for Computer Networks is conducted to prove the feasibility of our approach.It should be emphasized that what this paper has discussed is the relationship between Enlogy and FL. Research on the relationship between Enlogy and other kinds of Traditional Logic in the future is very important.