期刊名称:Journal of Theoretical and Applied Information Technology
印刷版ISSN:1992-8645
电子版ISSN:1817-3195
出版年度:2012
卷号:37
期号:2
页码:224-233
出版社:Journal of Theoretical and Applied
摘要:The sharp growth of online-systems and vast availability of high quality data lead to information overload, increasingly very difficult for the online users to find most relevant content. When looking for information about any movie, music, video, the internet users come across a bewildering number of options to fetch precise data from the recommended list. The main goal of the recommender system is to suggest high quality and top rated videos to the user. However there exist thousands of video items said to be Long Tail (videos with least rating) that stagnate idle on the web server for years that are unrevealed by users because of its least rating. The new recommender system introduced in this paper uses rating based binning technique that favors not only top rated videos to get recommended, but also recommend Long-Tail videos. This can improve diversity on recommendation and suggest best long tail videos to the user. This implies that a least popular video has a high probability to become more popular when it is placed on the related video recommendation lists of popular videos. In order to evaluate the proposed video recommender system, the datasets are crawled from YouTube�, a well-liked online video community to suggest videos with high rating along with less rated long tail videos.
关键词:Recommender System; Long Tail; Information Storage And Retrieval; Recommendation Diversity