Design And Implementation Of Mobile Internet User Preference Feature Analysis System

In recent years, with the rapid development of Mobile Internet, the Internet technology update and the widespread use of mobile terminal equipment, cell phone and Internet becomes the main way of living in the modern age. The application of mobile terminals records all kinds of preferences’ features of people in the Mobile Internet operation. It will bring new vitality and development to the enterprise, if they make analysis and use of these features. In addition, the enterprise not only need to seek self-innovative development, but also need to make strategic decisions according to the features of user, in order to design product which is more coherent to the user’s preferences. Even if mastering the user preferences, they can carry on accurate marketing via the most appropriate channels.Mobile Internet user preferences’ feature analysis system can fully use Mobile Internet access log data which directly reflects the features of Mobile Internet user preferences to analyze and mine data. Firstly, Mobile Internet user preferences using log parsing technology. Secondly, category user preferences using clustering algorithm based on a particular theme. Finally, Mobile Internet user preferences feature come to a conclusion through statistics and analysis. The data mining analysis system integrate the procedure of log collection, log parsing, user clustering and preferences analysis together.Firstly, the paper introduces the concept of Mobile Internet user preferences’ feature analysis system and the research status at home and abroad. On the basis of analyzing the core business process, the paper makes demand analysis of the system from two aspects of function and performance. Secondly, from the perspective of application, the system is divided into three subsystems as followed: Log analysis and statistic subsystem; User preferences analysis subsystem; Result show subsystem. The Log analysis and statistic subsystem provides data source which can be analyzed as the basis of user preferences’ analysis. User preferences analysis subsystem digs to get the user preferences analysis features using K-Means clustering algorithm. Result show subsystem represents the analysis results of preferences to data experts.The paper introduces a layered structure style, using the Hadoop distributed storage and computing framework. The paper carries on design and implementation in detail regarding each subsystem and module in the system. The content of functional design, network structure design, architecture design, database design and interface design is included. Finally, the paper tests on the function and performance requirements of the system in detail, in order to verify the availability of the system. The test result turns out the running condition of the Mobile Internet user preferences’ feature analysis system which is the topic of this paper meet the desired design expectation. The system has been applied by a mobile operator, running in a stable and reliable way.