| 摘要 |
<P>PROBLEM TO BE SOLVED: To recommend information in a field which is unexpected for a user and easy to be accepted by the user. <P>SOLUTION: A similarity calculation means 3b calculates a first index value expressing to what extent another user who belongs to one or more communities in common with a predetermined user treats information with affinity for the predetermined user, from characteristics of the communities to which they belong in common and characteristics of all the communities to which the predetermined user belongs, an unexpectedness calculation means 3c calculates a second index value expressing to what extent the other user who belongs to one or more communities in common with the predetermined user provides the information in the field which is unexpected for the predetermined user, from the characteristics of the communities to which they belong in common and characteristics of users who belong also to communities to which the predetermined user does not belong but other user belongs among users of the communities to which the other user belongs in common with the predetermined user, and a recommendation means 3d selects other user who recommends the information to the predetermined user based on results obtained by calculating the index values for each user. <P>COPYRIGHT: (C)2010,JPO&INPIT |
