SYSTEM AND METHOD FOR ITEM RECOMMENDATIONS

• US 20140081996A1
• Filed: 11/25/2013
• Published: 03/20/2014
• Est. Priority Date: 05/10/2001
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated
• Original Assignee: International Business Machines Corporation
• Inventors: BERTRAM, Ralf, SCHRIMPF, Gerhard, STAMM-WILBRANDT, Hermann
• Granted Patent: US 11,157,984 B2
• Time in Patent Office: Days
• US Class Current: 707/751
• CPC Class Codes: G06F 16/90324 using system suggestions
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0222 During e-commerce, i.e. online transactions
• CPC Class Codes: G06Q 30/0631 Item recommendations

Abstract

Means and a computerized method for recommending items such as books and audio compact disks. For each item, a user profile includes ratings provided by users of the system. Unlike present recommendation systems, the user profiles do not include pre-computed similarity factors measuring similarity between users. Rather, when an advisee requests a recommendation, similarity measures are computed comparing the advisee to other users, and the similarity measures are associated with the other users. A subset of the users is selected, where the subset includes the users most similar to the advisee. A recommendation is made based on the ratings by the members of the selected subset.

Claims

1. A recommendation system comprising a computer-readable storage device having program code stored in the storage device, wherein:

the computer readable storage device includes a plurality of records;

each of the plurality of records includes a user identifier, an item identifier, and a rating value;

each of the plurality of records is linked in a first and a second dimension;

the first dimension links records with a same user identifier in a sequence according to the item identifier; and

the second dimension links records with a same item identifier in a sequence according to the user identifier,

and the recommendation system is operable to;

update only similarity factors between neighboring users and an advisee for a new rating being entered for at least one advisee profile; and

determine a ranked matching list of the neighboring users.

2. A method for generating a recommendation of an item to an advisee, the method comprising:

receiving, by a computing apparatus of a recommendation system, a recommendation request including a selected item list of an advisee;

selecting, by the computing apparatus, a first set of users from a plurality of users of the recommendation system based on the selected item list;

selecting, by the computing apparatus, neighboring users from the first set of users based on similarities between the advisee and each member of the first set of users; and

generating, by the computing apparatus, a recommendation of at least one item from the selected item list based on ratings provided by each of the neighboring users.

3. The method of claim 2, wherein the recommendation of the at least one item includes only items that are included in the selected item list.

4. The method of claim 2, wherein the selecting the neighboring users comprises determining the similarities between the advisee and each member of the first set of users after receiving the selected item list.

5. The method of claim 4, wherein determining the similarities comprises:

determining a plurality of similarity factors based on at least one advisee profile; and

determining which of the plurality of users has already rated the at least one item.

6. The method of claim 2, further comprising determining the similarities by computing a plurality of similarity factors between the advisee and the neighboring users only for the items of the selected item list.

7. The method of claim 6, wherein all items upon which the computing depends are included in the selected item list.

8. The method of claim 6, wherein the plurality of similarity factors are not calculated with respect to all users of the recommendation system for every individual recommendation request.

9. The method of claim 2, further comprising assigning confidence factors to an advisee profile and user profiles of the first set of users, wherein the confidence factors are based on the combined effect of selected pieces of information recorded in the advisee profile and the user profiles.

10. The method of claim 2, wherein a member of the first set of users is selected as a neighboring user if the similarities between the advisee and the member are greater than a predetermined threshold.

11. The method of claim 2, further comprising assigning a weight to each of the neighboring users, wherein a first one of the neighboring users having a greater similarity to the advisee is assigned a weight that is greater than a second one of the neighboring users having a lesser similarity to the advisee.

12. The method of claim 2, further comprising updating the selected neighboring users each time a new rating is entered or inferred during the recommendation request.

13. The method of claim 2, further comprising creating a user profile for each of the plurality of users, wherein the user profiles do not include any pre-computed similarity factors.