| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To provide a recommendation information generation device and a recommendation information generation method, for properly selecting a similar user even if an amount of history information is small. <P>SOLUTION: In an information distribution server 200, a user property vector calculation part 204 calculates user property vectors of respective users, and a similar user calculation part 209 calculates similarity among the users based on the calculated user property vectors of the respective users. The similar user calculation part 209 selects the similar user similar to one user based on the calculated similarity, and a relevance evaluation part 210 generates recommendation information for the one user based on the property vector of the selected similar user. <P>COPYRIGHT: (C)2010,JPO&INPIT</p> |
