Collaborative filtering system and collaborative filtering method

A collaborative filtering system which uses evaluation values of contents from a plurality of users, the system comprising:

• a similarity calculation unit which, when there are no evaluation values from at least one user who has evaluated both a first content and a second content, calculates the similarity between the first content and the second content using evaluation values of a third content whose evaluation value is present from at least one user who has evaluated both the first content and the third content and whose evaluation value is present from at least user who has evaluated both the second content and the third content;
• and
• a predicted evaluation value calculation unit which calculates a predicted evaluation value from a user who has not evaluated either of the first content and the second content using the similarity between the first content and the second content calculated by the similarity calculation unit and the evaluation values of the first content and the second content,
• wherein, for one user who has used a specific content and has not evaluated the specific content, the predicted evaluation value calculation unit calculates a predicted evaluation value of the specific content from the one user using the use frequency of the specific content of the one user, the use frequency of the specific content of another user who has used the specific content and has evaluated the specific content, and an evaluation value of the specific content from another user.

A collaborative filtering method which uses evaluation values of contents from a plurality of users, the method comprising:

• a similarity calculation step of, when there are no evaluation values from at least one user who has evaluated both a first content and a second content, calculating the similarity between the first content and the second content using evaluation values of a third content whose evaluation value is present from at least one user who has evaluated both the first content and the third content and whose evaluation value is present from at least one user who has evaluated both the second content and the third content;
• and
• a predicted evaluation value calculation step of calculating a predicted evaluation value from a user who has not evaluated either of the first content and the second content using the similarity between the first content and the second content calculated in the similarity calculation step and the evaluation values of the first content and the second content,
• wherein, in the predicted evaluation value calculation step, for one user who has used a specific content and has not evaluated the specific content, a predicted evaluation value of the specific content from the one user is calculated using the use frequency of the specific content of the one user, the use frequency of the specific content of another user who has used the specific content and has evaluated the specific content, and an evaluation value of the specific content from another user.