The Research And Realization Of Personalized News Recommendation System Based On Mobile Internet

With the rapid popularizing and developing of the mobile Internet around theworld, a large amount of influx information has come into the vision of users who usemobile Internet in recent years, so people have entered into the "information overload"era of the mobile Internet from "lack of information" era. However, the informationof consumers and producers are meeting the great challenges. As to informationconsumers, how to choose the information in which they are interested is a difficulttask in the ocean of information. As to information producers, how to stand out oftheir own production, which is concern by the majority of users, but that, is also adifficult thing. Becoming the observers and practitioners to the field of computerscience has meeting unprecedented opportunities and challenges. It also becomes afocus and hot issues which is talking about in the Internet, especially in the field ofelectronic commerce and social field. Digging out the user’s behavior, characteristics,and understanding the user’s interest deeply, that will be bound to greatly enhance userstickiness and satisfaction after the analysis of the potential demand of the users.The personalized recommendation refers to the characteristics and buyingbehavior which is based on the user’s interest; it could recommend information andcommodities in which the users are interested, and it also could realize Internetinformation received from passive to active push. Collaborative-Filtering, which as arecommended Algorithm, is one of the most successful and most studied algorithms;that its basic idea is to be recommended. According to users interested in similarneighbors, then it will find out the largest user which is similar to the current user, andcalculate the neighbor users’ interest. It would generate recommendations after usingweighted as the current user interest.This article talks about the news in the mobile Internet system, which could help users improving the experience of reading the news on mobile terminal after studyingthe personalized recommendation in the field of the application. According to thetraditional recommendation, there are three improved recommended models in thisarticle, which are the content-based recommendation, the collaborative filtering basedon user attributes and similarity improvements recommendation and the improvedrecommendation based on the context of time. It could build the correlation betweenthe user and another user, and it would analyze the different time characteristics whichare used to recommend news to the user after quantifying and calculating the keywords of the news content. Based on these three recommended models, this article hasrealized a simple mobile Internet-based personalized news recommendation systemand finally passes the test. The user information and news information are processedand filtered, which is used to helping users finding news information in which theyare interested. Relative to the traditional news-type applications, this system hascombined the advantage of the traditional news with the personalized news.Firstly, this paper describes the background and significance of the topic, andthen describes the problem of personalized recommendation application status athome and abroad. Secondly, through analyzing the characteristics of the mobileInternet environment under News Recommended based on mobile Internet-basedpersonalized news system, it fulfills system detailed demand analysis, system design,database design-implementation, and the final completion of the testing of the entiresystem. Finally, this paper summarizes the full text of the proposed work and thenmaking an outlook for the next step.