Predicting interest of a user of a social networking service in a content item

Patent metadata

Field	Value
Publication number	US-9703837-B1
Application number	US-201314040424-A
Country	US
Kind code	B1
Filing date	Sep 27, 2013
Priority date	Sep 28, 2012
Publication date	Jul 11, 2017
Grant date	Jul 11, 2017

Abstract

The subject matter of this specification can be implemented in, among other things, a computer-implemented method including determining an affinity score representing an affinity of a user with respect to a contact of the user, wherein the affinity score is an indication of a strength of a relationship between the user and the contact, determining an engagement score of the user with respect to the contact, wherein the engagement score in an indication of a probability of the user engaging with a content item associated with the contact, determining an interest score of the user with respect to the contact based on the affinity score and the engagement score and providing a plurality of content items posted by the contact display in an activity stream, the content items being ordered based on the interest score. Other aspects can include corresponding systems, apparatus and computer program products.

First claim

What is claimed is:

1. A computer-implemented method comprising:
determining an affinity score representing an affinity of a user with respect to a contact of the user, wherein the affinity score is an indication of a strength of a relationship between the user and the contact;
determining an engagement score of the user with respect to the contact, wherein the engagement score is an indication of a probability of the user engaging with a content item associated with the contact;
determining affinity scores between the user and each participant of each of a plurality of content items posted by the contact, wherein participants of a content item comprise the contact and one or more of another user that interacts with the content item;
determining a post-level score for each of the plurality of content items posted by the contact, each post-level score based on the affinity scores between the user and each participant of the respective content item;
determining an interest score of the user with respect to the contact based on the affinity score and the engagement score; and
providing the plurality of content items posted by the contact for display in an activity stream, the content items being ordered based on the interest score or the respective post-level scores.

2. The method of claim 1 , wherein the engagement score is based on a number of opportunities for the user to read a first set of content items posted by the contact and a number of engagements by the user with the first set of content items.

3. The method of claim 2 , wherein the number of opportunities for the user to read the first set of content items posted by the contact and the number of engagements by the user with the first set of content items are both time-decayed numbers.

4. The method of claim 2 , wherein the engagements comprise one or more of an endorsement, a comment, a sharing of a post, a universal resource locator click, expansion of a content item or comment of the content item, or a prolonged read of the content item.

5. The method of claim 2 , wherein the first set of content items comprises content items posted by the contact displayed in a stream provided to the user.

6. The method of claim 1 , wherein the engagement score is based on a number of opportunities for one or more members of a social group to read a second set of content items posted by the contact and a number of engagements by the one or more members of the social group with the second set of content items, wherein the contact is a member of the social group.

7. The method of claim 6 , wherein the number of opportunities for one or more members of the social group to read the second set of content items posted by the contact is normalized to a maximum group reads number, and wherein the number of engagements by the one or more members of the social group with the second set of content items is normalized based on the maximum group reads number.

8. The method of claim 6 , wherein the number of opportunities for one or more members of the social group to read the second set of content items posted by the contact and the number of engagements by the one or more members of the social group with the second set of content items are both time-decayed numbers.

9. The method of claim 1 , wherein the engagement score is based on a number of opportunities for the user to read a third set of content items posted to the activity stream by a plurality of contacts and a number of engagements by the user with the third set of content items.

10. The method of claim 1 , wherein the affinity score is based on a time-decayed number of communication sessions between the user and the contact.

11. The method of claim 10 , wherein the communication sessions comprise one or more of electronic messages, text, audio, or video chat sessions, or telephone calls.

12. The method of claim 1 , wherein one or more of the engagement score or the affinity score is modified based on a relationship between the user and the contact.

13. The method of claim 1 , wherein one or more of either the engagement score or the affinity score is modified based on a communication between the user and the contact.

14. The method of claim 1 , further comprising:
determining an engagement score of the user with respect to each participant of the content item;
determining a post-level score of the content item based on the determined engagement score of the user with respect to each participant; and
ordering the content item in the activity stream based on the post-level score.

15. A computer-readable medium storing instructions that when executed cause a computer to perform operations, the operations comprising:
for each pair of a plurality of user and contact pairs:
determining an affinity score representing an affinity of the user with respect to the contact, wherein the affinity score is an indication of a strength of a relationship between the user and the contact;
determining an engagement score of the user with respect to the contact, wherein the engagement score is an indication of a probability of the user engaging with a content item posted by the contact; and
determining an interest score of the user with respect to the contact based on the affinity score and the engagement score;

receiving a request from a first user for content items for an activity stream;
retrieving a plurality of content items for display within the activity stream, each content item being posted by the contact of the first user;
determining a post-level score for each of the plurality of content items, each post-level score based on affinity scores or engagement scores between the first user and each participant of the respective content item, wherein participants of a content item comprise the contact and one or more of another user that interacts with the content item;
ordering the retrieved plurality of content items based on the determined interest scores of the user with respect to each of the contacts posting each of the plurality of content items or the respective post-level score for each of the plurality of content items; and
providing the ordered content items for display in the activity stream of the first user.

16. The computer-readable medium of claim 15 , wherein determining the interest score of the user with respect to the contact based on the affinity score and the engagement score further comprises:
selecting the greater of the affinity score and the engagement score as the interest score, when the affinity score is above an affinity score threshold; and
selecting the engagement score as the engagement score where the affinity score is below or equal to the affinity score threshold.

17. The computer-readable medium of claim 15 , wherein the engagement score is based on a number of opportunities for the user to read a first set of content items posted by the contact and a number of engagements by the user with the first set of content items.

18. The computer-readable medium of claim 17 , wherein the engagements comprise one or more of an endorsement, a comment, a sharing of a post, a universal resource locator click, expansion of a content item or comment of the content item, or a prolonged read of the content item.

19. The computer-readable medium of claim 15 , wherein the affinity score is based on a time-decayed number of communication sessions between the user and the contact.

20. A computer-implemented system, the system comprising:
one or more processors;
a memory comprising instructions which, when

Assignees

• Google Inc

Inventors

• Teng Junbin
• Tong Simon
• Mazniker Boris

Classifications

• G06Q10/40

  Business processes related to social networking or social networking services · CPC title

• G06Q30/0202 Primary

  Market predictions or forecasting for commercial activities · CPC title

• G06F17/3053 Primary

  Physics · mapped topic

• G06F16/24578 Primary

  using ranking · CPC title

• G06Q10/42

  Determination of affinities or common interests between users · CPC title