Systems and methods for providing improved recommendations

Patent metadata

Field	Value
Publication number	US-10475105-B1
Application number	US-201816035254-A
Country	US
Kind code	B1
Filing date	Jul 13, 2018
Priority date	Jul 13, 2018
Publication date	Nov 12, 2019
Grant date	Nov 12, 2019

Abstract

Systems and methods for providing improved recommendations are disclosed. In some embodiments, the systems and methods may be used for vehicle recommendations. The system may include a server system configured to receive user historical vehicle preferences, user vehicle preferences, generate weighted feature data sets, and apply a similarity model to the generated weighted feature data set in order to determine a vehicle recommendation data set. A visual representation of the vehicle recommendation data set may then be provided to an interface associated with a user.

First claim

We claim:

1. A method for providing improved vehicle recommendations comprising:
receiving, by a recommendation engine embodied in a server system, user vehicle preference data;
receiving, by the recommendation engine, user historical vehicle preference data;
generating, by the recommendation engine, a weighted feature data set based on the received user vehicle preference data and the received user historical vehicle preference data,
the generated weighted feature data set further comprising a plurality of data units,
applying, by the recommendation engine, a similarity model to the generated weighted feature data set to determine a vehicle recommendation data set including at least one vehicle and its associated vehicle features, wherein each vehicle feature is associated with a subset of the plurality of data units for the generated weighted feature data set;
applying, by the recommendation engine, a variability filter to the determined vehicle recommendation data set to generate a varied vehicle recommendation data set, wherein each vehicle in the vehicle recommendation data set comprises a vehicle make and a vehicle model, wherein the application of the variability filter to the vehicle recommendation data set ensures each vehicle in the varied vehicle recommendation set has a different vehicle make and vehicle model combination, wherein the varied vehicle recommendation data set comprises a subset of the determined vehicle recommendation data set; and
providing, by the recommendation engine, a visual representation of the varied vehicle recommendation data set for display in an interface associated with the user.

2. The method of claim 1 , wherein generating the weighted feature data set further comprises:
mapping the received user vehicle preference data to a user vehicle preference data set having data corresponding to the plurality of data units;
mapping the received user historical vehicle preference data to a user historical vehicle preference data set having data corresponding to the plurality of data units;
determining an influence factor for each of the received user vehicle preference data and the received user historical vehicle preference data; and
aggregating the user vehicle preference data set and the user historical vehicle preference data set in accordance with the determined influence factor to form the weighted feature data set.

3. The method of claim 1 , wherein the user vehicle preference data further comprises at least one of user preferences directly provided by a user corresponding to at least one of a vehicle and vehicle features, and user preferences directly provided by a user corresponding to a questionnaire related to at least one of vehicle features and a vehicle.

4. The method of claim 1 , wherein the user historical vehicle preference data further comprises user preferences for a vehicle determined based on historical viewing patterns of a user.

5. The method of claim 1 , wherein the similarity model is a cosine similarity model.

6. The method of claim 1 , wherein the vehicle features include at least one of fuel efficiency, mileage, price, engine, year, fuel type, drive train, exterior color, body style, condition, and transmission.

7. The method of claim 1 , wherein applying the similarity model to the generated weighted feature data set to determine a vehicle recommendation data set further comprises:
generating a feature representation for a vehicle data set;
determining one or more similarity values between the feature representation for the vehicle set and the generated weighted feature data set; and
selecting one or more vehicles corresponding to at least a subset of feature representation for the vehicle set having the closest determined similarity values.

8. A method for displaying improved vehicle recommendations comprising:
receiving user vehicle preference data from a user via a user interface;
determining user historical vehicle preference data based on user usage of the user interface;
transmitting the user vehicle preference data and the user historical vehicle preference data to a server system communicatively coupled to the user interface via a network,
wherein the server system applies a similarity model to determine a vehicle recommendation data set based at least in part on the user vehicle preference data and the user historical vehicle preference data, wherein the vehicle recommendation data set includes at least one vehicle and its associated vehicle features, and wherein each vehicle feature is associated with a subset of a plurality of data units for a generated weighted feature data set based on the user vehicle preference data and the user historical vehicle preference data, wherein the server system applies a variability filter to the determined vehicle recommendation data set to generate a varied vehicle recommendation data set, wherein each vehicle in the vehicle recommendation data set comprises a vehicle make and a vehicle model, wherein the application of the variability filter to the vehicle recommendation data set ensures each vehicle in the varied vehicle recommendation set has a different vehicle make and vehicle model combination, wherein the varied vehicle recommendation data set comprises a subset of the determined vehicle recommendation data set; and

receiving, from the server system, the varied vehicle recommendation data set; and
displaying, on the user interface, at least a portion of the varied vehicle recommendation data set.

9. The method of claim 8 , wherein the vehicle features includes at least one of fuel efficiency, mileage, price, engine, year, fuel type, drive train, exterior color, body style, condition and transmission.

10. The method of claim 8 , wherein the user vehicle preference data further comprises at least one of user preferences directly provided by a user corresponding to at least one of a vehicle and vehicle features, and user preferences directly provided by a user corresponding to a questionnaire related to at least one of vehicle features and a vehicle.

11. The method of claim 8 , wherein determining user historical vehicle preference data based on user usage of the user interface further comprises determining vehicles and or vehicle features previously browsed by the user.

12. A system for providing improved vehicle recommendations comprising:
a processor; and
non-volatile memory storing computer program code that when executed on the processor causes the processor to execute a process operable to:
receive user vehicle preference data;
receive user historical vehicle preference data;
generate a weighted feature data set based on the received user vehicle preference data and the received user historical preference data, the generated weighted feature data set further comprising a plurality of data units;
apply a similarity model to the generated weighted feature data set to determine a vehicle recommendation data set, wherein the vehicle recommendation data set comprises at least one vehicle and its associated vehicle features, wherein each vehicle feature is associated with a subset of the plurality of data units for the generated weighted feature data set;
apply a variability filter to the determined vehicle recommendation data set to generate a varied vehicle recommendation data set, wherein each vehicle in the vehicle recommendation data set comprises a vehicle make and a vehicle model, wherein the application of the variability filter to the vehicle recommendation data set ensures each vehicle in the varied vehicle recommendation set has a different vehicle make and vehicle model combination, wherein the varied vehicle recommendation data set c

Assignees

• Capital One Services Llc

Inventors

• Anderson Jacob
• Venuraju Mithra Kosur
• Mittal Shilpa
• Hoang Ben
• Deshpande Amit
• Shen Jie

Classifications

• G06Q30/0278 Primary

  Product appraisal · CPC title

• G06Q30/0631 Primary

  Recommending goods or services · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title

• G06F16/2457

  with adaptation to user needs · CPC title

• G06Q30/0643

  graphically representing goods, e.g. 3D product representation · CPC title