A location-based personalized news recommendation

Personalized news article recommendation provides interesting articles for a user based on the users' preference. With increasing use of hand-held devices the interests of users are reflected by their location. Therefore, we propose a novel method to incorporate the location into a user preference for the personalized news recommendation. The proposed method, a topic model based on LDA, represents the user preference with not only the news articles read by user but the locations of the user. By representing the user preference differently over her location, it suggests the news articles which are appropriate the user location. In the evaluation of the proposed method compared to the LDA, it outperforms in topic quality both quantitative and qualitative evaluation. In addition, the experimental result on news article recommendation shows that the location based user preference improves the performance of news article recommendation.