Abstract
Recommendation service is becoming popular in many e-commerce sites and content distribution sites (e.g. news, movies and games). In this service, the system identifies an individual by a login certification and extracts information about interest and preference from the user. It finally recommends items that a user seems to be interested in based on the information. However, a recommendation function is not popular in a TV set (recommending TV programs) whereas it is widely used in the Web-based services. The most crucial reason is that it is difficult to identify an individual sitting in front of the TV, which are usually shared by several members of the family. These users usually do not want to perform additional operation on the remote control for login authentication. This means that it is not possible to make a user profile about the personal interest and preference. Therefore, in this study, we propose a technique to individualize a user profile from the viewing log mixed with several users.
