The Technology Research Of Building Knowledge Base In The Field Of Electronic Commerce

Based on the rapid development of e-commerce,simple navigation,browsing and retrieval functions have been unable to meet the needs of clientele,and it has become a major concern of the electricity business that how to get useful information from a large number of data and make a reasonable decision on the foundation of the information to provide satisfactory service for customers.The construction of knowledge base is an effective measure to enhance the acquisition,sharing,maintenance and utilization of knowledge.As a knowledge-based system,knowledge base is beneficial to improve the quality of forecasting model and recommendation system,and it will be the core support for the enterprise to provide personalized,customized and intelligent service for consumers.At present,there are few researches on the construction of domain-based knowledge base,and the degree of automation of knowledge base construction is not high.In this paper,the concept lattice theory,ontology and knowledge engineering are used to improve the degree of automation of knowledge base construction.First of all,according to the knowledge base theory and the analysis of the current situation in the e-commerce field,this paper designs a knowledge base architecture oriented e-commerce domain.And then,the method of knowledge acquisition based on complete information and incomplete information is studied.Under the complete information,this paper constructs concept lattice by taking users’ personal information and commodity attribute information of e-commerce platform as data source based on concept lattice theory and domain ontology,using which to construct domain ontology and obtain association rules.Under the incomplete information,an approximate concept lattice is proposed to solve the problem of knowledge acquisition under imperfect information in the concept of interval concept lattice.By comparison with the upper/lower approximate concept lattice,the feasibility and effectiveness of the conceptual lattice is proved.Finally,the knowledge integration of merging of different ontologies and changing rules is researched.Merging of ontology means that Concept lattice construction and concept lattice merging of formal context formed by ontology analysis are finally transformed into ontology,while the rule updating is based on the change of the association rules generated in different time periods,and the change pattern of the rule and the method of identifying and judging are given,and the corresponding updating strategy is adopted according to the different rule change patterns.