| 摘要 |
<P>PROBLEM TO BE SOLVED: To provide a content selection system with which the search for recommended content, based on the content purchased or the like, by a user can be performed with a high degree of accuracy. <P>SOLUTION: A feature-quantity generating function 110 generates feature-quantity data of content with multi-dimension vector data by using content-related information, and stores the feature-quantity data. A learning result generating function section 120 generates learning result data indicative of a user's taste with multi-dimension vector data by executing clustering processing using the feature-quantity data of content selected based on the user's taste, and stores the learning result data. A content search function section 150 extracts the feature-quantity data the Euclidean distance of which the learning result data is within a predetermined distance from the feature-quantity data stored in the feature quantity generating function 110, and searches for the content. Search with a high degree of accuracy can be performed, due to the use of the feature-quantity data of the multi-dimension vector generated by the feature quantity generating function 110. <P>COPYRIGHT: (C)2011,JPO&INPIT |
