Research On Consumer Behavior Prediction Of Mobile Terminal Users Based On Context Awareness

With the development of mobile Internet and mobile computing in recent years, smart phones and other mobile devices become the main tools to communicate and exchange information. In this case, context-aware service appears and deeply affects our daily life. On one hand, context-aware computer is not only designed to capture the users’ current status and environmental information, but also to make “having what we think” possible. On the other hand, with the continuous improvement of our living standards, shopping has become an important part of our lives. If we can predict the potential needs of customers according to their current situation, it will be very meaningful to create a good and convenient shopping experience for customers: The prediction allows companies to have a higher accuracy in establishing their marketing strategy while to personalize their services. At the same time users can save time and energy.With various forms of consumption and “information overloading” brought by mobile Internet developing, how to obtain contextual information effectively and to predict consumer’s behavior based on its context has become a challenge in the field of personification service. Sponsored by the National Natural Science Foundation of China, this project studies user’s context information in mobile environments to capture and analyze the strategies, as well as to form methods based on consumer behavior. The object of this project is to provide a better user experience through mobile terminals with help of context-aware services.The main tasks are as follows:① Analyzing the current status of the domestic and international researches and looking for the solution of the problems existed.② Exploring theories and practices of context-aware service, and introducing two ways of obtaining the mobile user’s contextual information, one is direct context collection and the other is potential information collection.③ By model building and situational similarity algorithm, contextual information can be dealt with.④ Taking account of unique features of context awareness technology and mobile end-user consumer’s behavior, this paper provides some improvements for traditional user’s behavior analysis and prediction algorithms. It offers an algorithm which bases on contextual similarity, secondary clustering and collaborative filtering algorithm. The paper also carry out simulation experiments to verify its effectiveness and performance⑤ Combined with solutions from prediction of business mobile end-user consumer behavior, this paper designs a self-shopping guide system based on prediction of consumer behavior.