News personalized recommendation has long been a favourite domain for recommender research. Traditional approaches strive to satisfy the users by constructing the users' preference profiles. Naturally, most of recent methods use users' reading history (content-based) or access pattern (collaborative filtering based) to recommend proper news articles to them. Besides, some researches encapsule the news content and access pattern in a recommender by vector space model. In this paper, we propose to use a hypergraph ranking for obtaining the preference rough, and then utilize the binary decision tree for eliminating the definition subjectivity of hypergraph. In this way, we can combine the content attributes on news content attributes, users and user's access pattern in a unified hypergraph and get more accuracy results, whereas we needn't to construct the user profile and select the possible important attributes empirically. Finally, we designed several experiments compared to the state-of-the-art methods on a real world dataset, and the results demonstrate that our approach significantly improves the accuracy, diversity, and coverage metrics in mass data.
展开▼
