| 摘要 |
<P>PROBLEM TO BE SOLVED: To provide a recommender system, for automatically, efficiently and satisfactorily extracting an item to be recommended to a target user without requiring the user to provide explicit evaluation. <P>SOLUTION: In item correlation processing (S22), an item correlation value is calculated, for all items used by each user, based on a difference in use time between two items and information on the oldest use times of the items. In recommend result calculation processing (S23), a total item correlation value is calculated by totaling the item correlation values for all users, and an item not contained in the item use history of the target user, which has high total item correlation values with items contained in the item use history of the target user is calculated as a recommendation item. <P>COPYRIGHT: (C)2009,JPO&INPIT |
