Abstract:
| Many recommender systems are based on Content-based Filtering and Social Filtering. Both methods have their own advantages and disadvantages, and they complement each other rather than compete. So the incorporation of both methods can make a better system and the combination technique controls the quality of the entire recommender system. In this paper, we propose each user has his own tendency to decide which is the better recommendation for himself among the various recommendation results and suggest the personalized combination technique. To represent user tendency, we define and use loyalty, diversity, and pioneerity. We show by experiments that our combination technique is useful. This combination technique improved the average coverage by 23% and ceiling by 40%. |
Speaker Bio:
| Soojung Lee is a Lecturer and Instructor at Sejong University in Seoul, Korea. She received her MS in Computer Science and Engineering concentrating in Databases, Data Mining, Information Retrieval, and Recommendation from Seoul National University in February of 2003. As a student, she researched movie recommendation systems at SNU.s OOPSLA Lab. She has also served as a technical consultant to Films Inc. (Internet Movie Database Inc.). Her research interests include information filtering and recommendation, computational mechanisims for sociotechnology, information and communication technology, e-communities, and e-commerce. |