Machine learning modeling using social graph signals

Patent metadata

Field	Value
Publication number	US-11966853-B2
Application number	US-202117225767-A
Country	US
Kind code	B2
Filing date	Apr 8, 2021
Priority date	Jul 21, 2017
Publication date	Apr 23, 2024
Grant date	Apr 23, 2024

Abstract

Systems and methods are provided for receiving a request for lookalike data, the request for lookalike data comprising seed data and generating sample data from the seed data and from user data for a plurality of users, to use in a lookalike model training. The systems and methods further provide for capturing a snapshot of social graph data for a plurality of users and computing social graph features based on the seed data and the user data for the plurality of users, training a lookalike model based on the sample data and the computed social graph features to generate a trained lookalike model, generating a lookalike score for each user of the plurality of users in the user data using the trained lookalike model, and generating a list comprising a unique identifier for each user of the plurality of users and an associated lookalike score for each unique identifier.

First claim

What is claimed is:

1. A method comprising:
receiving, at a server computer system, a request for lookalike data, the request for lookalike data comprising seed data;
generating, by the server computer system, sample data from the seed data and from user data for a plurality of users, to use in a lookalike model training, by performing operations comprising:
generating a positive data sample from the seed data to use in the lookalike model training; and
generating a negative data sample from user data stored in a database to use in the lookalike model training;

training, by the server computer system, a lookalike model based on the sample data and user profile features for the plurality of users to generate a trained lookalike model;
generating, by the server computer system, a lookalike score for each user of the plurality of users in the user data using the trained lookalike model;
capturing, by the server computer system, a snapshot of social graph data for the plurality of users and determining at least one social graph feature;
generating, by the server computer system, a social graph score for each user of the plurality of users by comparing each user of the plurality of users to each user of the seed data based on the at least one social graph feature;
performing, by the server computer system, optimized ranking of the plurality of users based on the generated lookalike score for each user and the social graph score for each user;
generating, by the server computer system, a final lookalike score for each user of the plurality of users based on the optimized ranking; and
generating, by the server computer system, a list comprising a unique identifier for each user of the plurality of users and an associated final lookalike score for each unique identifier.

2. The method of claim 1 , wherein the positive data sample comprises the seed data.

3. The method of claim 1 , wherein the negative data sample comprises a subset of the user data for the plurality of users.

4. The method of claim 1 , further comprising:
capturing a user profile snapshot;
generating user profile feature data; and
storing the user profile feature data.

5. The method of claim 1 , wherein the request further comprises filter characteristics, and the method further comprises:
filtering the list comprising each user and an associated final lookalike score for each user, based on filter characteristics received in the request for lookalike data; and
associating the filtered list with the request for lookalike data.

6. The method of claim 1 , wherein the seed data comprises a plurality of user identifiers.

7. The method of claim 1 , wherein the generated list comprising a unique identifier for each user of the plurality of users and an associated final lookalike score for each unique identifier, is of a size indicated by the request for the lookalike data.

8. The method of claim 1 , wherein the plurality of users are users of a messaging system or social networking system.

9. The method of claim 1 , wherein the generated list is associated with a requester that sent the request for the lookalike data, and the method further comprises:
receiving content from the requester; and
displaying the content to one or more users of the plurality of users, based on the generated list.

10. A server computer comprising:
one or more hardware processors; and
a computer-readable medium coupled with the one or more hardware processors, the computer-readable medium comprising instructions stored thereon that are executable by the one or more hardware processors to cause the server computer to perform operations comprising:
receiving a request for lookalike data, the request for lookalike data comprising seed data;
generating sample data from the seed data and from user data for a plurality of users, to use in a lookalike model training, by performing operations comprising:
generating a positive data sample from the seed data to use in the lookalike model training; and
generating a negative data sample from user data stored in a database to use in the lookalike model training;
training a lookalike model based on the sample data and user profile features for the plurality of users to generate a trained lookalike model;
generating a lookalike score for each user of the plurality of users in the user data using the trained lookalike model;
capturing a snapshot of social graph data for the plurality of users and determining at least one social graph feature;
generating a social graph score for each user of the plurality of users by comparing each user of the plurality of users to each user of the seed data based on the at least one social graph feature;
performing optimized ranking of the plurality of users based on the generated lookalike score for each user and the social graph score for each user;
generating a final lookalike score for each user of the plurality of users based on the optimized ranking; and
generating a list comprising a unique identifier for each user of the plurality of users and an associated final lookalike score for each unique identifier.

11. The server computer of claim 10 , wherein the positive data sample comprises the seed data.

12. The server computer of claim 10 , wherein the negative data sample comprises a subset of the user data for the plurality of users.

13. The server computer of claim 10 , the operations further comprising:
capturing a user profile snapshot;
generating user profile feature data; and
storing the user profile feature data.

14. The server computer of claim 10 , wherein the request further comprises filter characteristics, and the operations further comprise:
filtering the list comprising each user and an associated final lookalike score for each user, based on filter characteristics received in the request for lookalike data; and
associating the filtered list with the request for lookalike data.

15. The server computer of claim 10 , wherein the seed data comprises a plurality of user identifiers.

16. The server computer of claim 10 , wherein the generated list comprising a unique identifier for each user of the plurality of users and an associated final lookalike score for each unique identifier, is of a size indicated by the request for the lookalike data.

17. The server computer of claim 10 , wherein the plurality of users are users of a messaging system or social networking system.

18. The server computer of claim 10 , wherein the generated list is associated with a requester that sent the request for the lookalike data, and the operations further comprising:
receiving content from the requester; and
displaying the content to one or more users of the plurality of users, based on the generated list.

19. A non-transitory computer-readable medium comprising instructions stored thereon that are executable by at least one processor to cause a computing device to perform operations comprising:
receiving a request for lookalike data, the request for lookalike data comprising seed data;
generating sample data from the seed data and from user data for a plurality of users, to use in a lookalike model training, by performing operations comprising:
generating a positive data sample from the seed data to use in the lookalike model training; and
generating a negative data sample from user data stored in a database to use in the lookalike model training;
training a lookalike model based on the sample data and user profile features for the plurality of users to generate a trained lookali

Assignees

• Snap Inc

Inventors

• Blackwood John Cain
• Brewer Jason
• Khajehnouri Nima
• Minooei Hadi
• Steele Benjamin C
• You Qian

Classifications

• G06N5/022 Primary

  Knowledge engineering; Knowledge acquisition · CPC title

• G06F16/951

  Indexing; Web crawling techniques · CPC title

• G06N20/00 Primary

  Machine learning · CPC title

• G06F16/9536

  Search customisation based on social or collaborative filtering · CPC title