A Mobile Internet Business Recommendation Model Based On Data Analysis Of User Behavior

With the rapid development of domestic Mobile Internet industry and triple play in recent years, market competition between the products of Mobile Internet is sent more intense. For operators, the competition between the Value Added Service will become the competition between the service qualities. As the products of Mobile Internet are so multifarious,domestic operators began a integration Customer-Centric operation instead of the production separate operation. New business moles and intense with each passing day competition environment put forward some new requirements and challenges to telecom value-added business operation management.The main aim of this article is to analysis the data form behaviors of Mobile Internet users, foreseen the user demand tendency, accurate positioning appropriate product to user and promotion business. Further the objects of"Form customers find business, for business find customers".In this article, we have first discussed the user behavior analysis and methods; learn kinds of algorithm and software of data mining. On this basis, we construct models of user behavioral analysis and business matching, understand and collect data form Mobile Internet. According to the ETL (Extraction-Transformation-Loading)means, we make up,sort and improve the 70 million records that we got and build server of database. Form the hot time,user interests,matching business perspectives, we make knowledge discovery in databases using methods of cluster,text mining,association analysis. We find marketing strategic issues form the statistical data, and provide the support for new measures of marketing and precision marketing of operators.The system builed in this paper has a concise and clear design idea, which has been proved effective. It provides good foundation for achieving following system of analysis users with more data sources and greater data quantity.