Design And Implementation Of A News Recommendation System Integrating User Clustering And Collaborative Filtering

Since the 21st century,the rapid development of the Internet has brought a lot of convenience to people’s lives.A large amount of data is generated in the form of self-media and spread by a large number of users.The arrival of the era of big data has provided users with a wealth of resource information.People have access to resource information from traditional newspapers,radio and other media to obtain the information they need from the network.It is very convenient to read online news on the Internet.People get an important source of information,but it also brings many problems.The huge amount of news resources makes it difficult for users to obtain the required news information in a large amount of news data,which causes the problem of information overload.On the other hand,many users do not have clear news information when they browse the news.Therefore,collecting real and accurate news in a large amount of news data,and making corresponding news recommendations according to the user’s potential interest,to meet the needs of users,can bring great social and economic value.This paper takes the design and implementation of news recommendation system integrating user clustering and collaborative filtering as the research topic.The main research contents are as follows:(1)Research background and significance of the research,research on related theories and techniques involved in this paper,including data acquisition technology,clustering algorithm and recommendation technology.(2)On the basis of relevant theoretical techniques,research on news recommendation techniques that integrate user clustering and collaborative filtering.In this paper,based on the cold start and data sparseness in the recommendation system,the recommendation technology of fusion clustering is adopted.Based on the existing improved clustering recommendation technology,the recommendation strategy for new users is added,from different user perspectives.Use different methods for recommendations.The new users are clustered according to the registration information,and then collaborative filtering recommendation;the old users are clustered according to the historical behavior and then recommended.Compared with the traditional algorithm,this algorithm has a good recommendation effect.(3)Analyze and design the overall function of the system and each module.The functional modules of the system include news data collection module,user clustering module and news recommendation display module.The news data collection module uses the Scrapy crawler framework for data collection,crawling the data of major news websites and storing them in a structured form;the user clustering module uses different clusters for new and old users of the system;the news recommendation module is in the user After clustering,a collaborative filtering recommendation algorithm is used in the category to generate news recommendations.(4)A news recommendation system that integrates user clustering and collaborative filtering to test the function and performance of the system.After testing,the system is fully functional,and can quickly and efficiently collect news data at regular intervals and make accurate news for different users,recommend.