Recommendation system with user interface for exposing downstream effects of particular rating actions

• US 8,478,664 B1
• Filed: 10/25/2011
• Issued: 07/02/2013
• Est. Priority Date: 10/25/2011
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Xavier, Joseph, Ionkov, George M., Lehman, Jeffrey D., Doshi, Japan S., Qiu, Sandy Wenling, Nath, Bobby P.
• Primary Examiner(s): STOPP, COURTNEY L
• Application Number:
  US13/280,593
• Time in Patent Office: 616 Days
• Field of Search: 705/26.1, 705/26.7, 705/26.8, 705/27.1
• US Class Current: 705/26.7
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/0631 Item recommendations

Abstract

An interactive system generates personalized item recommendations for users based partly or wholly on ratings assigned by the users to particular items. The system includes an item rating user interface that enables a user to view, prior to rating an item, information regarding the types of items that will be recommended to the user if the user assigns a particular rating or type of rating to the item. The user interface thereby enables users to refrain from performing rating actions that will tend to result in low utility or “poor quality” recommendations from the users'"'"' perspectives.

Claims

1. An interactive system, comprising:

a server that provides network-based access to an electronic catalog of items, said server providing a user interface that includes functionality for users to rate individual catalog items; and

a recommendation service that generates personalized recommendations of catalog items for users based at least partly on the ratings assigned by the users to particular catalog items;

said server configured to generate, for a user, a personalized page that provides functionality for the user to rate at least one catalog item, and to view, prior to rating the catalog item, information regarding items that would be recommended to the user by the recommendation service as a result of the user favorably rating the item, the personalized page thereby enabling the user to assess a downstream effect of favorably rating the catalog item before rating the catalog item.

2. The interactive system of claim 1, wherein the information regarding items that would be recommended includes a list of items that are representative of types of items that will be recommended if the user favorably rates the item.

3. The interactive system of claim 2, further comprising a personalization component that generates the list of items by a process that comprises:

looking up, from a pre-generated data repository of item-to-item association mappings, a set of catalog items that are behaviorally associated with the catalog item; and

filtering out, from the set of behaviorally associated items, one or more catalog items that, based on recorded information associated with the user, are already known to the user.

4. The interactive system of claim 2, further comprising a personalization component that generates the list of items by a process that comprises:

generating, with the recommendation service, a first set of item recommendations based on an existing profile of the user;

generating, with the recommendation service, a second set of item recommendations based on an augmented version of said profile, said augmented version reflecting a hypothetical favorable rating of the catalog item by the user; and

determining, by comparing the first and second sets of item recommendations, one or more catalog items that are included in the second set of item recommendations but not in the first set of item recommendations.

5. The method of claim 1, wherein the information regarding items that would be recommended to the user identifies item attributes characteristic of types of items that would be recommended.

6. The method of claim 1, wherein the personalized page, when presented on a user computing device, is responsive to use selection of the catalog item by displaying said information in a popover.

7. A computer-implemented method of notifying a user of a downstream effect of an item rating action, the method comprising:

providing a user interface that includes functionality for users to rate individual catalog items of an electronic catalog;

providing a recommendation service that generates personalized recommendations of items for users based at least partly on the ratings assigned by the users to particular catalog items;

determining, by a computer system, a set of items that are representative of types of items that would be recommended for the user by the recommendation service as a result of the user favorably rating a particular catalog item; and

notifying the user, via said user interface, of said set of items that are representative of types of items that would be recommended for the user as a result of the user favorably rating the particular catalog item, to thereby enable the user to assess a downstream effect of favorably rating the particular catalog item before favorably rating the particular catalog item.

8. The method of claim 7, wherein determining the set of items comprises:

looking up, from a pre-generated data repository of item-to-item association mappings, a set of items that are behaviorally associated with the particular catalog item; and

filtering out, from the set of behaviorally associated items, one or more items that, based on recorded information associated with the user, are already known to the user.

9. The method of claim 7, wherein determining the set of items comprises personalizing the set of items for the user based on recorded information reflective of item preferences of the user.

10. The method of claim 7, wherein determining the set of items comprises:

generating, with the recommendation service, a first set of item recommendations based on an existing profile of the user;

generating, with the recommendation service, a second set of item recommendations based on an augmented version of said profile, said augmented version reflecting a hypothetical favorable rating of the item by the user; and

determining, by comparing the first and second sets of item recommendations, one or more items that are included in the second set of item recommendations but not the first set of item recommendations.

11. The method of claim 7, wherein notifying the user comprises generating, and transmitting to a user computing device, a page that is responsive to user selection of the particular catalog item by displaying a popover that lists the determined set of items.

12. The method of claim 7, wherein the method is performed by an interactive computer system that includes the recommendation service, and wherein the set of items is determined using item selection criteria used by the recommendation service.