Wechat Official Accounts Article Recommendation System Based On User Preferences

In recent years,with the rapid development of Internet and intelligent terminal,social applications has come to us faster and faster.WeChat,as one of the most widely used applications on mobile terminals,has become an important way for people to obtain information.However,the mass data also brings the problem of information overload to us.How can the user obtain valuable information from numbers of Wechat official accounts is worthy of study.To solve the problem,this paper is on the basis of collecting and analyzing the content of Wechat official accounts article.Design Wechat official accounts article recommendation system using user preference to provide users with more accurate recommendations.The main work of this paper includes the following aspects:1.Put forward a new Wechat official accounts article crawling method.By designing the Wechat official accounts article crawling system,includes automatic access module,network package sniffing module and data parsing and storage module to automatically crawl data.This system can obtain reading number and thumb up number,that can't get in traditional way.It is efficient,stable and real-time.2.About the work of data preprocessing,we use Ansj to segmentate word,count word frequency and extract keyword.We have improved the keywords extraction algorithm,so that it can automatically load the local stop word dictionary and extract keyword more accurate3.We have designed and realized a Wechat official accounts article recommendation system,which can recommend articles in two ways.On content-based recommendation,it can find similar articles by calculating the similarity of word frequency vector,and analyze user preference to recommend articles.On tag-based recommendation,it can recommend articles related to user tags.4.After using hybrid recommendation,we overcame the shortcomings of content-based recommendation and tag-based recommendation and improved the precision and recall index of the system.The experimental results show that the recommendation system designed and implemented in this paper can crawl Wechat official accounts articles efficiently and steadily and recommend articles accurately.Compared to the traditional content-based recommendation,the system has better recommendation effect,which has certain practical application value.