Negative signals for advertisement targeting

Patent metadata

Field	Value
Publication number	US-10380606-B2
Application number	US-201213566016-A
Country	US
Kind code	B2
Filing date	Aug 3, 2012
Priority date	Aug 3, 2012
Publication date	Aug 13, 2019
Grant date	Aug 13, 2019

Abstract

Users of a social networking system perform actions on various objects maintained by the social networking system. Some of these actions may indicate that the user has a negative sentiment for an object. To make use of this negative sentiment when providing content to the user, when the social networking system determines a user performs an action on an object, the social networking system identifies topics associated with the object and associates the negative sentiment with one or more of the topics. This association between one or more topics and negative sentiment may be used to decrease the likelihood that the social networking system presents content associated with a topic that is associated with a negative sentiment of the user.

First claim

What is claimed is:

1. A computer-implemented method comprising:
storing, on a server, a user profile for a first user of a social networking system;
receiving, by the server, one or more actions performed by the first user on a first object maintained by the social networking system;
identifying, by multiple processors of the server, a topic associated with the first object, wherein the one or more actions performed by the first user on the first object subsequent to the first object being displayed to the first user indicate that the first user has a positive sentiment for the first object, but the one or more actions performed by the first user on the first object do not indicate whether the first user has a negative sentiment or a positive sentiment for the topic associated with the first object based on a context of use of the topic in the first object;
retrieving, from the server, one or more actions previously performed by second users of the social networking system on one or more second objects that are also associated with the topic of the first object, the second users having user profiles indicating a negative sentiment for the topic associated with the first object that received the one or more actions by the first user subsequent to the first object being displayed to the first user that indicate the positive sentiment of the first user for the first object, the negative sentiment of the second users for the topic based on the one or more actions performed by the second users on the second objects;
determining, by the multiple processors of the server, that the second users previously performed one or more actions on the first object that received the one or more actions by the first user subsequent to the first object being displayed to the first user that indicate the positive sentiment of the first user for the first object;
inferring, by the multiple processors of the server, that the first user has a negative sentiment for the topic associated with the first object even though the one or more actions performed by the first user on the first object subsequent to the first object being displayed to the first user indicate that the first user has a positive sentiment for the first object, the inference based on the second users previously performing one or more actions on the first object that received the one or more actions by the first user, the second users previously performing one or more actions on the one or more second objects that are also associated with the topic of the first object, and the second users having user profiles indicating negative sentiment for the topic associated with the first object;
storing, on the server, the topic as a negative interest in connection with the user profile; selecting, by the multiple processors of the server, content for presentation to the first user based at least in part of the negative interest;
aggregating, by the multiple processors of the server, the selected content for presentation to the first user in a continuously updated real-time newsfeed;
configuring, by the multiple processors of the server, the continuously updated real-time newsfeed into a format for presenting on a device of the first user; and
presenting the configured continuously updated real-time newsfeed on the device of the first user.

2. The method of claim 1 , wherein inferring that the first user has a negative sentiment for the topic comprises:
inferring, by the multiple processors of the server, the first user has the negative sentiment for the topic responsive to determining that at least one of the actions performed by the first user matches an action previously performed on at least one of the second objects associated with the topic by at least one of the second users of the social networking system having a user profile indicating the negative sentiment for the topic associated with the first object.

3. The method of claim 1 , wherein inferring that the first user has a negative sentiment for the topic comprises:
inferring, by the multiple processors of the server, the first user has the negative sentiment for the topic responsive to determining that at least a threshold number of the actions performed by the first user match actions previously performed on one or more second objects associated with the topic by the second users of the social networking system having user profiles indicating the negative sentiment for the topic associated with the first object.

4. The method of claim 1 , wherein the actions previously performed on one or more second objects associated with the topic by second users of the social networking system having user profiles indicating the negative sentiment for the topic associated with the first object comprise at least one selected from a group consisting of: closing a second object associated with the topic, hiding the second object associated with the topic, ignoring the second object associated with the topic, unliking the second object associated with the topic, not sending a response to the second object associated with the topic within a specified time interval and providing text input about the second object associated with the topic that includes one or more words associated with negative sentiment.

5. The method of claim 1 , further comprising:
identifying, by the multiple processors of the server, one or more third objects on which the first user performed one or more actions, the one or more third objects associated with one or more topics matching the one or more topics associated with the first object; and
associating, by the multiple processors of the server, the negative sentiment with the one or more third objects based on the actions performed on the one or more third objects.

6. The method of claim 1 , wherein the first object comprises at least one selected from the group consisting of: advertisements, posts, videos, images, stories, events and groups.

7. A computer-implemented method comprising:
receiving, by a server, actions performed on objects by a user of a social networking system that indicate a negative sentiment towards the objects, each of the objects associated with a plurality of topics;
identifying, by multiple processors of the server, a topic from the plurality of topics that is common to all of the objects;
determining, by the multiple processors of the server, a total number of actions performed on the objects by the user that indicate the negative sentiment towards the objects;
responsive to the total number of actions associated with negative sentiment equaling or exceeding a threshold, associating, by the multiple processors of the server, the negative sentiment with the topic common to all of the objects;
selecting, by the multiple processors of the server, content for presentation to the user who performed the actions based on the negative sentiment for the topic
aggregating, by the multiple processors of the server, the selected content for presentation to the user in a continuously updated real-time newsfeed;
configuring, by the multiple processors of the server, the continuously updated real-time newsfeed into a format for presenting on a device of the user; and
presenting the configured continuously updated real-time newsfeed on the device of the user.

8. The method of claim 7 , wherein identifying a negative sentiment for the objects based on the actions on the objects comprises:
determining, by the multiple processors of the server, whether one or more actions performed on an object are actions associated with negative sentiment by the social networking system; and
responsive to determining at least one action performed on the object is an action associated with the negative sentiment by the social networking system,

Assignees

• Garcia Martinez Antonio Felipe
• Facebook Inc

Inventors

• Garcia-Martinez Antonio Felipe

Classifications

• G06Q30/0269 Primary

  based on user profile or attribute · CPC title

• G06Q30/02 Primary

  Marketing; Price estimation or determination; Fundraising · CPC title

• G06Q10/40

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

• G06Q50/01

  Physics · mapped topic

• G06Q10/42

  Determination of affinities or common interests between users · CPC title