User Interest Analysis with Hidden Topic in News Recommendation System

To take advantage of the Internet - vast but complicated information resources, Recommendation systems help users find out information they need by providing them personalized suggestions. This research area is receiving more and more attention from researchers and used in some famous websites like EBay, Amazon, etc. In this paper, we proposed a Recommendation System for Vietnamese electronic newspaper which uses content-based filtering techniques associating with the attention of users shown in user´s profile. These users´ attentions are determined by inferring a set of common Hidden Topics from the documents which users preferred. Experimental results showed that approach is feasible with positive results and its capabilities for reality development.