Design And Implementation Of APP Personalized Recommendation System

With the development of mobile communication technology,smart mobile terminal has become an indispensable part of people's life.The increase of mobile users brings huge business profits to enterprises.Mobile application,as an important identification of mobile terminal,is the link that enterprises provide services for users.With this demand,the number of various mobile applications has grown exponentially.Faced with massive applications,it is difficult for users to find applications that they are interested in.Some application markets try to solve this problem by using classification and retrieval methods,but it turns out that user's personalized experience can not be improved in this way.Although some application markets have introduced some recommendation technologies,the problem of data sparsity,user interest migration and scalability is still serious.In this paper,aiming at the problem of data sparsity and user interest migration in the current mobile application market,after researching the related theoretical model of collaborative filtering recommendation algorithm and mobile user interest model,a collaborative filtering recommendation algorithm based on user interest model is proposed and verified.On this basis,a personalized mobile application recommendation system is designed and implemented based on the big data platform.This paper has accomplished the following tasks:The current situation at home and abroad of personalized recommendation system and mobile application recommendation is studied.The problems faced by the current mobile recommendation system are described.The theories of personalized recommendation technology and user interest model are introduced.The characteristics of the mobile application itself and its recommendation characteristics are analyzed.According to its characteristics of high privatization,dynamicity,and sparse scoring,a collaborative filtering recommendation algorithm based on user interest model is proposed and verified.Based on the big data platform,a mobile personalized application recommendation system is implemented,which includes a data acquisition module,a user modeling module,a candidate set generation module,and an online recommendation module.Finally,the functional test and performance test of the implemented system are carried out.The test result shows that the system's functional modules and the system's real-time performance and online recommendation performance can meet the expected requirements..