Development of a Recommender System based on Extending Contexts of Content and Personal History

摘要

—The flood of information on the Internet makes a person who approaches it without some strong intention feel overwhelmed. One way to redress the balance between a person and the flood is a computer-based recommender system, and many Web sites use such systems. These systems on a Web site work for similar items. However, the field of personal activity is not limited to handling one kind of knowledge or one Web site, but also involves off-line activities in the real world. To handle personal off-line activity, LifeLog was proposed as a method to record it, but the main purpose of LifeLog is to record a personal history. The uses of such a history are still being studied. We have developed a recommender system that captures personal context from a history of personal online and offline activities, treats information on Web sites as a large set of context, and discovers and extends the overlaps of personal activities and Web sites, then recommends information located in the Web sites. The aim of the system is to allow a person to enjoy waves of information again. The system was implemented as part of the My Life Assist Service for mobile phones provided by NTT DOCOMO, Inc. as a field experiment from December 2007 to February 2008.