Research On Semantic Recommended Method Of Agricultural E-commerce

Personalized recommendation is a technology based on user preferences and buying behavior to recommend interesting information and commodity o users.In recent years,it gradually become a hot research topic,which is widely used in electronic commerce,movies,music,video,social networking,reading,personalized messages and advertising etc.Personalized recommendation can solve the problem of "information overload",and can meet the individual needs of users in different background to a certain extent.However,with the development of e-commerce of agricultural products,agricultural products electronic business website system have tens of millions of commodity information,the traditional recommendation methods can not do well in the user feature recommending,resulting in the low utilization rate of information.Therefore,in order to more objectively and accurately and quickly to recommend the service to users,it is necessary to study the recommended methods of agricultural e-commerce semantic.This research aimed at the problem of low accuracy of recommendation in agricultural product electronic commerce system.Use Dabieshan products e-commerce as the object of study,focused on solving based on two key problems of the ontology semantic recommendation,user interest modeling the semantics of problem,agricultural products related to semantic recommendation problem and proposed recommendation method based on semantic agricultural electronic business intelligence.Research on the construction method of user model based on Ontology projection algorithm.Propose A semantic recommendation method based on similarity degree and correlation degree.And develop agricultural e-commerce semantic recommendation system.The main contents and achievements of this thesis are as follows:(1)The construction method of user model based on Ontology projection algorithm is studied,which solves the problem of semantic modeling of user's interest.Through database extraction in commodity attribute and its characteristic value,properties and characteristics of sampling value for processing and using the Web Ontology Language(OWL)unified knowledge description method to manually construct of tea domain ontology,make full use of domain ontology concepts,attributes and instance description of the user interest,from the semantic interpretation of user interest,and finally the use of projection algorithm to generate the user personalized ontology.(2)An intelligent recommendation method based on semantic similarity and correlation is proposed to solve the problem of semantic recommendation of agricultural products.By calculating the taxonomic and non taxonomic relations in the ontology,using semantic similarity and semanticcorrelation algorithm to calculate the concept of semantic distance and correlation degree in the user model and the were clustering and interest degree forecast,finally get the conclusion that the recent neighbor users in order to complete the recommendation.(3)Develop the web client and Android client products in Dabeshan agricultural e-commerce semantic recommendation system,the background using the Java language,in the Eclipse platform development completed Dabieshan products of intelligent electronic commerce recommendation system,and verify the correctness and feasibility of the theory and method of.The research results to improve cognitive information meet the personalized needs of users,improve the electronic commerce system cross selling ability,enhance the customer loyalty of e-commerce enterprise and greatly enhance corporate earnings has a important significance.