Analysis Of Mobile Interent User Behavior Based On Interents

With the rapid development of mobile Internet technology,more and more users visit the Internet by smart device,so that mobile Internet has become one of the most important ways for users to access the Internet.Compared with the traditional PC Internet,Internet access through mobile device is more convenient and flexible.The proliferation of mobile terminal makes it possible for users to browse what they prefer anytime and anywhere.What's more important is that a variety of mobile Internet applications related to all aspects of our daily life are emerging in an endless stream,which has been fully penetrated into people's daily lives and has been changing the way of living and interactive for human.Under this background,a good understand of Mobile user's Behavior based on interests makes it easier to satisfy their demands,improves the quality of internet service,and even reduces the cost of time for information filtering.In this thesis,mobile Internet data from Dec.28th 2013 to Jan.3rd 2014 is collected,which is obtained from the Mobile Internet of a city in the south of China.On account of the data,mobile use's behavior based on interests is analyzed from three aspects.Initially,user interest community is discovered by using spectral cluster with Nystrom.The method shows efficiency and applicability.Meanwhile,the distribution of traffic for each communities of interest is also described.Secondly,improved Apriori algorithm is applied to find the association rule and estimate the strength of correlation between user's location and application.According to the association rule we found,behavior with strong correlation between the two features are extracted.Finally,one of the time series model,Hidden Markov Model(HMM)is applied to model the behavior we extracted.Based on the model constructed,we predict application individual prefer to use in a given location.