Research And Implementation Of Mobile News Recommendation Based On Context Awareness

In the era of mobile Internet,with the increasing popularity of mobile devices such as mobile phones,users can read news conveniently anytime and anywhere.The news recommendation algorithm can quickly find the news that matches the reading interest and improve the reading experience of users.By using mobile devices to collect abundant contextual information and using contextual information to improve the accuracy and diversity of recommendation results,this thesis studies user-interest modeling and news recommendation algorithms based on context awareness,including the following aspects:1.For the problems of density peak clustering algorithm,this thesis proposes a new density peak clustering algorithm based on KNN.The algorithm first uses the KNN to calculate the local density of the sample,and then selects the initial cluster centers by the least-squares linear fitting and assigns the remaining samples to form the initial clustering results.Finally,clustering the initial clustering results based on the density reachability analysis to generate the final clustering result.Experimental results on cluster dataset and sougou text dataset validate the effectiveness of the proposed method.2.Aiming at the demand of quick response of mobile news recommendation,this thesis proposes a simple and efficient mobile news recommendation algorithm based on Naive Bayes.The algorithm first uses the naive Bayesian model to obtain the user's preference of news categories in different contexts,and then scoring the candidate news according to the popularity of the news and the time of publication,and finally selects the highest rated N news as the news recommendation result.The experimental results on the actual data set verify the feasibility and effectiveness of the proposed algorithm.3.Based on the above research results,this thesis developed a mobile news recommendation system.The system is divided into server and client.The server realizes the collection and processing of mobile news data,the expression of contextual user interest and the generation of news recommendation results.The client is a mobile news application for Android,it provides read news,post comments and other functions,used to test mobile users' satisfaction with recommended results online.