| With the rapid development of mobile communication technology, the network bandwidth of mobile communication is increasing continuously, the third generation mobile communications business emerges, so the user of mobile streaming media becomes more and more, the processing ability of mobile terminal is also in increasing quickly, providing streaming media service through the mobile communication network has become a fact. The number of mobile users around the world is very huge, by the end of2011, the number of global mobile user has reached5.9billion. According to Ericsson’s research, it shows that the mobile streaming media service has great market, which has become one of the hot spots of the global mobile business research, because the global daily adds2million mobile users.This paper researches the patching preferred algorithm, streaming media proxy server cache technology based on the video-on-demand system, user behavior analysis, RFM model, improved incremental mining based on weight and mobile platform technology. According to the characteristics of streaming media, improves the traditional proxy caching strategy, and provides personalized optional recommendation streaming media. Through the realization of streaming media proxy server cache strategy based on the patching preferred algorithm, it optimizes the proxy server cache. At the same time, we put forward to a personalized streaming media recommendation system based on the user behavior analysis and incremental mining based on weight algorithm, combining with users recent habit and behavior, and putting forward to incremental mining based on weight algorithm, to increase the amount of data incremental to mining recently rules, mining association rules by the Apriori algorithm. And calculating all users similarity with similar vector matrix, similar gathered themselves together. Finally, using the collaborative filtering method, recommendation module will recommend the video to users, as a personalized recommendation way to complete the entire recommendation process. It can improve the users’ dependency and love on streaming media client with high efficiency.Finally, we implements a mobile streaming media client based on Android platform, the client can play mobile streaming media well, and at the same time, it can provide personalized recommendation according to different users. |