| With the rapid increase of the number of mobile users and the development of mobile technology, mobile information service including weather forecasting and news service is becoming more and more popular.Meanwhile, different users have different requests. Some users often fall into a dilemma because they often face a lot of useless information when they try to search some specific information. Therefore, personalized information service system which can meet users' demands effectively needs to be established and should be implemented.As a thesis, this article introduces a new model of mobile information service system in order to match the need of mobile users. The function of this model is to provide personalized news recommending. This article involves current research about this topic, building the model of the system and building the model of mobile users. And concerned algorithms about personalized recommendation are also purposed. Besides, the article emphasizes on recommending valuable information for mobile users.In this article, after introducing the significance of this research topic and other backgrounds, this article describes a new mobile information service system based on off-line information browsing system and Agent theories. Some basic principles of off-line information browsing system, data mining and personalized recommendation are introduced at first. Some details of modeling users, predicting users' interests, and personalized recommending are discussed and the association between users' behavior and their true interests are analyzed. Then a detailed algorithm for finding users' interests and some methods to implement each module are purposed. At last, concerned experiments and analysis are described. Conclude and future works are also introduced at the end of this article. |
PDF Full Text Request