Research On Mobile E-Commerce Recommendation System Based On Granular Computing

With the fast development of mobile e-commerce, the rapid popularization of intelligent terminals, social networks, along with extended field of information technology, cloud computing theory has been proposed, the rise of semantics and ontology, people already entered the era of big data. However, with the amazing amount of information, people have to pay more cost in order to find the right they need in a large number of flood information. But with regard to information retrieval, processing and display, etc. Compared with the traditional PC, mobile terminal has a significant weakness. How to analysis of user needs efficiently, so that users are not overwhelmed by large information and to help users found the information or resources they're interest efficiently and accurately has become one of the hottest topics in the research of mobile e-commerce recommendation system. At the same time, how to better adaptation characteristic of mobile e-commerce mobility, new features virtual, personalized, extreme data, and social features, also become the key content of mobile e-commerce recommendation system research.Granular computing is a new research field in recent years, the rise, the information processing of a new concept and computing paradigm is mainly used for description and uncertain, fuzzy, incomplete, and a mass of information processing and provide a kind of granular and intergranular based on the relationship between problem solving methods. The basic idea of granular computing is to use granular in solving problem. It describes and solves the problem of the abstract, the relations between granular, synthesize?decomposition and transactions of granular The model of granular computing is consistent with the structure of the real world, people's thinking mode and behavior mode. It provides an understanding, generalization, and operation of the problem by multiple levels and multiple perspectives. A new idea and method for the research of mobile e-commerce recommendation system model is provided by the thought of granular.Based on the mobile e-commerce recommendation system model as the research object, granularity analysis cited mobile e-commerce recommendation model, according to the characteristics of mobile e-commerce, with recommendation model of existing research methods and means, using the granularity analysis of granulation, synthesis, delamination, make full use of the granularity analysis based on structured thinking philosophy and methodology based on the structured problem solving method, the calculation mode based on structured information processing, multi levels and multi view structure of thought introduced to mobile e-commerce recommendation system model, construct a personalized recommendation system based on granular computing model.For mobile e-commerce recommendation based on the granular analysis, we use the Dianping.com Chengdu station catering data to experimental analysis the personalized recommendation model. From the aspect of the result set and the properties of granular to analysis recommendation algorithm accuracy and diversity. Our researches are expected not only to give a new way to solve the research difficulties of mobile e-commerce recommendation system in theory and method, but also can expand the application field of granular computing theory, has the important theory and application value.