System and method for recommending items

• US 4,996,642 A
• Filed: 09/25/1989
• Issued: 02/26/1991
• Est. Priority Date: 10/01/1987
• Status: Expired due to Term

• Current Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Original Assignee: Neonics, Inc.
• Inventors: Hey, John B.
• Primary Examiner(s): Lall, P. S.
• Assistant Examiner(s): MELNICK, STEVEN A
• Application Number:
  US07/411,857
• Time in Patent Office: 519 Days
• Field of Search: 364/401, 364/419, 364/410, 364/554, 364/550, 434/237, 434/236, 358/86, 358/84, 455/2, 273/161
• US Class Current: 705/7.29
• CPC Class Codes: G06F 16/30 of unstructured textual data document management systems G06F16/93
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G09B 5/00 Electrically-operated educational appliances working with questions and answers G09B7/00; simulators G09B9/00; advertising or displaying in general G09F
• CPC Class Codes: H04N 21/466 Learning process for intelligent management, e.g. learning user preferences for recommending movies services using the results of monitoring in broadcast systems H04H60/61

Abstract

A system and method of selectively recommending, for a user selected from a group of users, items such as movies sampled by one or more of the users in the group but not sampled by the selected user. The recommendations are based on other items previously sampled by the user and preferably on the availability of the items from a source, for example the availability of movies from a video store. A scalar rating is defined for each item sampled by the selected user to represent the reaction of the selected user to that item. The selected user is successively paired with the users in a group for whom have been defined scalar ratings for at least some of the items sampled by the selected user to determine the difference in ratings for items sampled by both members of that pair. An agreement scalar is generated for each pair representing the overall rating agreement between the members of each successive pair. A plurality of the other users are designated as recommending users and the agreement scalar is converted into a weighting value for each of the recommending users. The weighting values are applied to items not yet sampled by the selected user to adjust the recommendation for the selected user to more closely anticipate the actual reaction of the user to that item. Preferably, at least one item is selected to be presented to the selected user based on a predetermined criterion and on the availability of the item from the source.

Claims

1. A method of selectively recommending, for a user selected from a group of users, items sampled by one or more users in the group but not sampled by the selected user, the recommendations being based on other items previously sampled by that user and on the availability of the items from a source, the method comprising:

defining, for each item sampled by the selected user, a scalar rating representing the reaction of the selected user to that item;

successively pairing the selected user with other users in the group for whom have been defined scalar ratings for at least some of the items sampled by the selected user to determine the difference in ratings for items sampled by both members of that pair, and generating for each pair an agreement scalar representing the overall rating agreement between the members of each successive pair;

designating a plurality of the other users as recommending users and converting, for each of the recommending users, the agreement scalar into a weighting value;

applying the weighting values to items not yet sampled by the selected user to proportionally alter the difference between a rating previously established for each item not yet sampled by the selected user and the ratings of that item by the recommending users to adjust the recommendations for the selected user to more closely anticipate the actual reaction of the user to that item; and

selecting at least one item to be presented to the selected user based on a predetermined criterion and on the availability of the item from the source.

2. The method of claim 1 in which pairing includes successively matching, for each pair, items sampled by both members and, for each matched item, comparing the ratings of one member from the rating of the other member to obtain the difference in ratings.

3. The method of claim 2 in which pairing further includes converting, for each pair, the difference in ratings for each matched item to a closeness value, and combining the closeness values for the members of that pair.

4. The method of claim 3 in which pairing further includes weighting, for each pair, the combined closeness values by the number of items sampled by both members to generate the agreement scalar.

5. The method of claim 3 in which designating and converting includes defining a greater weighting value for recommending users having a larger agreement scalar and defining a lesser weighting value for recommending users having a smaller agreement scalar.

6. The method of claim 3 in which designating and converting includes ranking the recommending users by ascending order of agreement scalar and in that order assigning to the ranked recommending users progressively larger weighting values.

7. The method of claim 1 in which the predetermined criterion is selection of at least the most highly recommended item for the selected user.

8. The method of claim 1 in which the predetermined criterion is specified by the user prior to the recommendation.

9. The method of claim 1 in which applying includes combining the weighting value for each recommending user with the difference between the actual rating by that recommending user and the previously recommended rating by the selected user for each unsampled item, and summing the combination with the previously recommended rating.

10. The method of claim 1 in which pairing includes successively pairing the selected user with each other user in the group.

11. The method of claim 1 in which pairing includes defining a subset of other users in the group to be successively paired with the selected user.

12. The method of claim 1 in which selecting includes determining the availability of the item by interrogating a system which maintains an inventory of items at the source.

13. The method of claim 1 further including maintaining an inventory status of the items at the source, and selecting includes determining the availability of the item by comparing the item with the inventory status for that item.

14. The method of claim 1 further including successively selecting the remainder of the users in the group to adjust the recommendations for each user in the group.

15. The method of claim 1 further including presenting the selected item to the selected user.

16. A method of selectively recommending, for each user successively selected from a group of users, items sampled by one or more users in the group but not sampled by the selected user, the recommendations being based on other items previously sampled by that user and on the availability of the items from a source, the recommendations being represented by a scalar rating for each item, the method comprising:

defining, for each item sampled by the selected user, a scalar rating representing the reaction of the selected user to that item;

successively pairing the selected user with other users in the group for whom have been designated scalar ratings for at least some of the items sampled by the selected user to determine the difference in rating for items sampled by both members of each successive pair;

generating for each pair an agreement scalar representing the overall rating agreement between the members of that pair;

designating at least one of the other users as recommending users;

converting, for each of the recommending users, the associated agreement scalar into a recommendation fraction;

identifying items unsampled by the selected user;

applying the recommendation fractions to proportionally decrease the difference between a rating previously established for each identified item for the selected user and the actual ratings of that item by the recommending users to adjust the recommendations for the selected user;

selecting at least one item to be presented to the selected user based on a predetermined criterion and on the availability of the item from the source;

successively selecting the remainder of the users in the group to adjust the recommendations for each user in the group; and

presenting, for each user, a plurality of items selected for that user.

17. The method of claim 16 in which pairing includes successively matching, for each pair, items sampled by both members and, for each matched item, comparing the rating of one member with the rating of the other member to obtain the difference in ratings.

18. The method of claim 17 in which pairing further includes converting, for each pair, the difference in ratings for each matched item to a closeness value, and combining the closeness values for the members of that pair.

19. The method of claim 18 in which pairing further includes weighting, for each pair, the combined closeness values by the number of items sampled by both members to generate the agreement scalar.

20. The method of claim 19 in which designating and converting includes sorting the recommending users by ascending order of agreement scalar and in that order assigning to the ranked predicting users progressively larger weighting values.

21. The method of claim 20 in which applying includes combining the weighting fraction for each recommending user with the difference between the actual rating by that recommending user and the previous predicted rating by the selected user for each identified item, and summing the combination with the previously established rating.

22. The method of claim 21 further including initially setting the established rating for each unsampled item for each selected user to a low rating value.

23. The method of claim 16 in which selecting includes determining the availability of the item by interrogating a system which maintains an inventory of items at the source.

24. The method of claim 16 further including maintaining an inventory status of the items at the source, and selecting includes determining the availability of the item by comparing the item with the inventory status for that item.

25. A system for selectively recommending, for a user selected from a group of users, items sampled by one or more users in the group but not sampled by the selected user, the recommendations being based on other items previously sampled by that user and on the availability of the items from a source, the system comprising:

means for defining, for each item sampled by the selected user, a scalar rating representing the reaction of the selected user to that item, said means for defining including input means for entering information representing the reaction of the selected user to items sampled by that user;

means for successively pairing the selected user with other users in the group for whom have been defined scalar ratings for at least some of the items sampled by the selected user to determine the difference in ratings for items sampled by both members of each successive pair;

means for designating a plurality of the other users as recommending users and assigning a weighting value to each of the recommending users based on the overall difference in ratings between that predicting user and the selected user;

means for applying the weighting values to items not yet sampled by the selected user to proportionally alter the difference between a rating previously established for each item not yet sampled by the selected user and the ratings of that item by the recommending users to adjust the recommendations for the selected user to more closely anticipate the actual reaction of the user to that item; and

means for selecting at least One item to be presented to the selected user based on a predetermined criterion and on the availability of the item from the source.

26. A computing device for selectively recommending, for a user selected from a group of users, items sampled by one or more users in the group but not sampled by the selected user, the recommendations being based on other items previously sampled by that user and on the availability of the items from a source, the system comprising:

means for defining, for each item sampled by the selected user, a scalar rating representing the reaction of the selected user to that item, said means for defining including input means for entering information representing the reaction of the selected user to items sampled by that user;

means for successively pairing the selected user with other users in the group for whom have been defined scalar ratings for at least some of the items sampled by the selected user to determine the difference in rating for items sampled by both members of each successive pair, said means for pairing including means for generating for each pair an agreement scalar representing the overall rating difference between the members of that pair;

means for designating a plurality of the other users as recommending users and for converting, for each of the recommending users, the

agreement scalar into a weighting fraction;

means for establishing an initial scalar rating for each identified item for the selected user;

means for identifying items not yet sampled by the selected user and applying the weighting values to items not yet sampled by the selected user to proportionally alter the difference between one of the initial scalar rating for each identified item and a rating previously predicted for each identified item and the ratings of that item by the predicting users to adjust the recommendations for the selected user to more closely anticipate the actual reaction of the user to that item; and

means for selecting at least one item to be presented to the selected user based on a predetermined criterion and on the availability of the item from the source.

27. A method of selectively recommending, for a viewer selected from a group of viewers, movies sampled by one or more viewers in the group but not sampled by the selected viewer, the recommendations being based on other movies previously sampled by that viewer, the method comprising:

defining, for each movie sampled by the selected viewer, a scalar rating representing the reaction of the selected viewer to that movie;

successively pairing the selected viewer with other viewers in the group for whom have been defined scalar ratings for at least some of the movies sampled by the selected viewer to determine the difference in ratings for movies sampled by both members of that pair, and generating for each pair an agreement scalar representing the overall rating agreement between the members of each successive pair;

designating a plurality of the other viewers as recommending viewers and converting, for each of the recommending viewers, the agreement scalar into a weighting value; and

applying the weighting values to movies not yet unsampled by the selected viewer to proportionally alter the difference between a rating previously established for each movie not yet sampled by the selected viewer and the ratings of that movie by the recommending viewers to adjust the recommendations for the selected viewer to more closely anticipate the actual reaction of the viewer to that movie.

28. A computing device for selectively recommending, for a viewer selected from a group of viewers, movies sampled by one or more viewers in the group but not sampled by the selected viewer, the recommendations being based on other movies previously sampled by that viewer and on the availability of the items from a source, the system comprising:

means for defining, for each movie sampled by the selected viewer, a scalar rating representing the reaction of the selected viewer to that movie, said means for defining including input means for entering information representing the reaction of the selected viewer to movies sampled by that viewer;

means for successively pairing the selected viewer with other viewers in the group for whom have been defined scalar ratings for at least some of the movies sampled by the selected viewer to determine the difference in rating for movies sampled by both members of each successive pair, said means for pairing including means for generating for each pair an agreement scalar representing the overall rating difference between the members of that pair;

means for designating a plurality of the other viewers as recommending viewers and for converting, for each of the recommending viewers, the agreement scalar into a weighting fraction;

means for establishing an initial scalar rating for each identified movie for the selected viewer; and

means for identifying movies not yet sampled by the selected viewer and for applying the weighting values to movies not yet sampled by the selected viewer to proportionally alter the difference between one of the initial scalar rating for each identified movie and a rating previously predicted for each identified movie and the ratings of that movie by the predicting viewers to adjust the recommendations for the selected viewer to more closely anticipate the actual reaction of the viewer to that movie.