On the Internet, many websites are offering question and answer system. These websites are known as knowledge sharing platforms. Knowledge sharing platforms allow users to ask and answer questions. Users may add tags to support easy categorization of their questions. Adding tags to the questions makes contents organized and easy to filter. However, tagging also has few issues such as tag explosion and tag synonym. Tag explosion represent there exist a common tag for the most of questions. Tag synonym represents there exist too many synonyms for a single tag. In this paper, an efficient probabilistic contents tag recommendation system is proposed. The proposed system topic model techniques based on probabilistic to resolve issues on knowledge sharing platforms. Proposed system utilizes advantages of topic models, and considers the contents similarity and word appearance to overcome these challenges.
Volume 11 | 05-Special Issue
Pages: 207-211