Research On Mobile Internet Personalized Information Recommendation Based On J2ME

With the popularity of Mobile Internet applications, Mobile Internet is bound to appear the phenomenon of "information flood", "information overload" etc. This paper designs a J2ME based mobile personalized recommendation model to address this problem, and analysis and implements the correlative algorithms.This paper first introduces the current status of Mobile Internet, and researches the technology of personalized information recommendation of Mobile Internet. On this basis, this paper designs a mobile personalized recommendation model based on the combination of ATC and CF. For the ATC technology used in the model, this paper uses the weighted naive Bayes (WNB) Classifier; for the problems of current CF algorithms, this paper designs and implements a collaborative filtering algorithm based on user classification and records credibility weighted method. Secondly, this paper implements the server side program of the model with JAVA and Servlet, and implements the client side software using J2ME technology. The experimental results show that the model can more accurately recommend information for users, and achieve a satisfied performance in efficiency and accuracy. This design of model has both theoretical and practical significance for solving the information overload on Mobile Internet.