Granted Patent B1

US 10,127,596 · App. 14/566,402 · Granted Nov 13, 2018

Inventors: David Wayne Franke (Austin, TX); Joshua Howard Levy (Austin, TX); Hans Ulrich Grasemann (Austin, TX); Lauri Janet Moore (Austin, TX); David Pratt (Austin, TX); William T. Moose (Austin, TX); Andrew K. Moore (Austin, TX); John William Prior (Austin, TX)

Assignee: VAST.COM, INC.

G06Q30/0631G06Q30/0627G06Q30/0641

Abstract

Claims (26)

1. A system for generating user specific recommendations of alternative unique items, the system comprising:

one or more computer readable storage devices configured to store a plurality of computer executable instructions; and

one or more hardware computer processors in communication with the one or more computer readable storage devices to execute the plurality of computer executable instructions in order to cause the system to:

render on a plurality of user devices an interactive electronic interface for presenting unique item listings and enabling a user to search for and interact with the unique item listings, wherein the unique item listings are associated with unique items each comprising a plurality of attributes including at least a condition attribute and a location attribute;

monitor over a computer network the plurality of user devices to detect user interactions with the unique item listings presented by the interactive electronic interface;

generate and store in an electronic database historical user preference data related to the detected interactions of the user with the unique item listings presented by the interactive electronic interface,

wherein the stored historical user preference data comprises at least a preference indicator for each of a plurality of unique items associated with unique item listings interacted with by the user;

analyze the stored historical user preference data to generate a user specific scoring model based at least in part on attribute differences between unique items preferred by the user and unique items not preferred by the user,

wherein generating the user specific scoring model comprises at least:

creating training data comprising a plurality of pairs of unique items that were detected as interacted with by the user, each of the plurality of pairs of unique items comprising difference data indicative of differences between one or more condition attributes and one or more location attributes;

labeling each of the plurality of pairs with the preference indicator associated with one of the unique items of each pair; and

inputting the training data into a supervised learning algorithm to generate a function that can output a predicted relative level of preference for an alternative unique item based on differences between condition and location attributes of the alternative unique item and a selected unique item;

receive, via the interactive electronic interface, a selection of a unique item by the user;

access, in an electronic database, attribute data related to a plurality of alternative unique items, the attribute data comprising at least condition attribute data and location attribute data;

generate a recommendation score for each of the plurality of alternative unique items indicating the predicted relative level of preference by the user in the alternative unique items, wherein the recommendation score for each alternative unique item is generated by at least:

analyzing the condition attribute data and location attribute data to determine differences between the attribute data of each alternative unique item and attribute data of the selected unique item; and

inputting the determined differences into the user specific scoring model to output the predicted relative level of preference; and

in response to the selection of the unique item by the user, re-render the interactive electronic interface to present the selected unique item and at least a subset of the alternative unique items sorted by their recommendation scores;

detect, by the monitoring of the plurality of user devices, additional user interactions by the user with the re-rendered interactive electronic interface;

update the historical user preference data based on the detected additional user interactions;

regenerate the user specific scoring model using the updated historical user preference data; and

responsive to the user specific scoring model being regenerated, re-render the interactive electronic interface to present at least a subset of the alternative unique items sorted by new recommendation scores generated using the regenerated user specific scoring model.

2. The system for generating user specific recommendation of alternative unique items of claim 1 , wherein the historical user preference data comprises the plurality of pairs of unique items that were detected as interacted with by the user, wherein one item in a pair of items is preferred by the user and the other item in a pair of items is not preferred by the user.

3. The system for generating user specific recommendations of alternative unique items of claim 2 , wherein the user specific scoring model is further generated using characteristics of the user received by the system.

4. The system for generating user specific recommendations of alternative unique items of claim 1 , wherein the detected interactions of the users with the unique item listings comprise one or more of the following interactions: likes, favorites, purchases, adds to a wish list, follows, or views details.

5. The system for generating user specific recommendations of alternative unique items of claim 1 , wherein the unique items associated with the unique item listings comprise one or more of used vehicles or existing homes.

Assignments (1)

Continuity (4)

Cited By (10)

US 12,254,111 US 12,288,175 US 12,321,974 US 12,326,910 US 12,367,517 US 12,367,518 US 12,481,723 US 12,536,574 US 12,579,568 US 12,731,183