The Analysis Of Mobile Phone Game Players Behavior Based On Data Mining

In the rapid development information age of mobile Internet,mobile phone game users are constantly expanding,mobile phone games usher in a huge market.In addition,with the further growth of the mobile game industry market size,the charge mode of mobile phone games and the consumption tendency of game players are gradually beyond many traditional industries.In this context,major mobile game operators on the game player's maintenance and development is facing increasing challenges.The use of data mining technology to conduct in-depth analysis of mobile phone game player behavior data,and access to potential information that is conducive to business operational decisions,which help mobile game operators to maintain competitive advantage.Therefore,thers is important theoretical significance and practical value to use data mining technology to analyze mobile phone game player behavior.This paper first analyzes the domestic and international research status of customer segmentation,personalized recommendation and mobile game player behavior.According to the shortcomings of the current research,this paper puts forward the research contents and methods.Secondly,this paper briefly introduces the basic theory of customer behavior analysis,the data mining algorithms involved in this paper and so on.Moreover,aiming at the characteristics of mobile phone game player behavior,this paper puts forward the evaluation index system of mobile gamer lifetime value by improving the traditional RFM value evaluation model,which considering the mobile game gamer's login value,the paid value and the game character value.Meanwhile this paper build a new three-dimensional player segmentation model based on the evaluation index system,and through the principal component analysis method to get a more accurate player value score,which in order to carry out targeted resource allocation strategy for each segment player group.Then,aiming at the client game player,the collaborative filtering recommendation algorithm is improved by introducing the player's subspace interest difference similarity,so as to build a personalized game recommendation model,which in order to enhance the level of mobile phone game player relationship management,maintain the activity of game players,and maximally keep the mobile game player traffic.Finally,each model is applied to the mobile phone game player behavior of a certain provincial China Mobile.The results show that the new three-dimensional player segmentation model can better identify the different players groups,status and values,and the personalized game recommendation model can improve the scalability of collaborative filtering recommendation algorithm to a certain extent,and improve the data sparsity influence on the player's mobile game recommended quality.