SYSTEMS AND METHODS FOR IDENTIFYING AND GROUPING RELATED CONTENT LABELS

• US 20170053013A1
• Filed: 08/18/2015
• Published: 02/23/2017
• Est. Priority Date: 08/18/2015
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Weinsberg, Ehud, State, Bogdan, Herdagdelen, Ama, Xiao, Bai, Torres de Sa Resende, Danilo, Dimson, Thomas Frederick
• Granted Patent: US 10,296,634 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/38 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Systems, methods, and non-transitory computer-readable media can acquire a set of labels associated with a set of content items. Each label in the set of labels can be associated with at least one content item in the set of content items. It can be determined that at least two labels, out of the set of labels, are related. The at least two labels can be determined to be related based on at least one of a co-occurrence metric associated with the at least two labels or a topic similarity metric associated with the at least two labels. One label can be selected, out of the at least two labels, as being representative of the at least two labels.

Claims

1. A computer-implemented method comprising:

acquiring, by a computing system, a set of labels associated with a set of content items, each label in the set of labels being associated with at least one content item in the set of content items;

determining, by the computing system, that at least two labels, out of the set of labels, are related, the at least two labels being determined to be related based on at least one of a co-occurrence metric associated with the at least two labels or a topic similarity metric associated with the at least two labels; and

selecting, by the computing system, one label, out of the at least two labels, as being representative of the at least two labels.

2. The computer-implemented method of claim 1, wherein the at least two labels are determined to be related based on the co-occurrence metric, and wherein determining that the at least two labels are related further comprises:

identifying each particular content item, out of the set of content items, that is associated with the at least two labels;

incrementing, for each particular content item that is associated with the at least two labels, the co-occurrence metric associated with the at least two labels; and

determining that the co-occurrence metric at least meets a specified co-occurrence threshold.

3. The computer-implemented method of claim 1, wherein the at least two labels are determined to be related based on the co-occurrence metric, and wherein determining that the at least two labels are related further comprises:

determining a number of times in which a first label and a second label, out of the at least two labels, are associated with a node in a social graph of a social networking system;

incrementing, based on the number of times, the co-occurrence metric associated with the at least two labels; and

determining that the co-occurrence metric at least meets a specified co-occurrence threshold.

4. The computer-implemented method of claim 3, wherein the node is associated with at least one of a particular label, a particular content item, a particular entity, a particular page, a particular group, a particular event, or a particular place.

5. The computer-implemented method of claim 1, wherein the at least two labels are determined to be related based on the topic similarity metric, and wherein determining that the at least two labels are related further comprises:

acquiring textual information associated with the set of content items;

determining, based on the textual information, a set of topic distributions for the set of content items, the set of topic distributions including a first topic distribution for a first content item in the set of content items and a second topic distribution for a second content item in the set of content items;

associating the first topic distribution with a first label out of the at least two labels and the second topic distribution with a second label out of the at least two labels, the first label being descriptive of the first content item and the second label being descriptive of the second content item;

calculating the topic similarity metric based on comparing the first topic distribution and the second topic distribution; and

determining that the topic similarity metric at least meets a specified topic similarity threshold.

6. The computer-implemented method of claim 5, wherein the textual information associated with the set of content items includes at least one of a respective caption for each content item in the set of content items or a respective description for each content item in the set of content items.

7. The computer-implemented method of claim 1, wherein the one label is selected based on having a highest social engagement metric with respect to the at least two labels.

8. The computer-implemented method of claim 1, further comprising:

providing a suggestion to utilize the one label.

9. The computer-implemented method of claim 1, wherein the set of labels includes a set of hashtags.

10. The computer-implemented method of claim 9, wherein the set of hashtags are determined to be trending with respect to at least one of a specified time period or a specified recent hashtag amount.

11. A system comprising:

at least one processor; and

a memory storing instructions that, when executed by the at least one processor, cause the system to perform;

acquiring a set of labels associated with a set of content items, each label in the set of labels being associated with at least one content item in the set of content items;

determining that at least two labels, out of the set of labels, are related, the at least two labels being determined to be related based on at least one of a co-occurrence metric associated with the at least two labels or a topic similarity metric associated with the at least two labels; and

selecting one label, out of the at least two labels, as being representative of the at least two labels.

12. The system of claim 11, wherein the at least two labels are determined to be related based on the co-occurrence metric, and wherein determining that the at least two labels are related further comprises:

identifying each particular content item, out of the set of content items, that is associated with the at least two labels;

incrementing, for each particular content item that is associated with the at least two labels, the co-occurrence metric associated with the at least two labels; and

determining that the co-occurrence metric at least meets a specified co-occurrence threshold.

13. The system of claim 11, wherein the at least two labels are determined to be related based on the co-occurrence metric, and wherein determining that the at least two labels are related further comprises:

determining a number of times in which a first label and a second label, out of the at least two labels, are associated with a node in a social graph of a social networking system;

incrementing, based on the number of times, the co-occurrence metric associated with the at least two labels; and

determining that the co-occurrence metric at least meets a specified co-occurrence threshold.

14. The system of claim 11, wherein the at least two labels are determined to be related based on the topic similarity metric, and wherein determining that the at least two labels are related further comprises:

acquiring textual information associated with the set of content items;

determining, based on the textual information, a set of topic distributions for the set of content items, the set of topic distributions including a first topic distribution for a first content item in the set of content items and a second topic distribution for a second content item in the set of content items;

associating the first topic distribution with a first label out of the at least two labels and the second topic distribution with a second label out of the at least two labels, the first label being descriptive of the first content item and the second label being descriptive of the second content item;

calculating the topic similarity metric based on comparing the first topic distribution and the second topic distribution; and

determining that the topic similarity metric at least meets a specified topic similarity threshold.

15. The system of claim 11, wherein the set of labels includes a set of hashtags.

16. A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computing system, cause the computing system to perform a method comprising:

acquiring a set of labels associated with a set of content items, each label in the set of labels being associated with at least one content item in the set of content items;

determining that at least two labels, out of the set of labels, are related, the at least two labels being determined to be related based on at least one of a co-occurrence metric associated with the at least two labels or a topic similarity metric associated with the at least two labels; and

selecting one label, out of the at least two labels, as being representative of the at least two labels.

17. The non-transitory computer-readable storage medium of claim 16, wherein the at least two labels are determined to be related based on the co-occurrence metric, and wherein determining that the at least two labels are related further comprises:

identifying each particular content item, out of the set of content items, that is associated with the at least two labels;

incrementing, for each particular content item that is associated with the at least two labels, the co-occurrence metric associated with the at least two labels; and

determining that the co-occurrence metric at least meets a specified co-occurrence threshold.

18. The non-transitory computer-readable storage medium of claim 16, wherein the at least two labels are determined to be related based on the co-occurrence metric, and wherein determining that the at least two labels are related further comprises:

determining a number of times in which a first label and a second label, out of the at least two labels, are associated with a node in a social graph of a social networking system;

incrementing, based on the number of times, the co-occurrence metric associated with the at least two labels; and

determining that the co-occurrence metric at least meets a specified co-occurrence threshold.

19. The non-transitory computer-readable storage medium of claim 16, wherein the at least two labels are determined to be related based on the topic similarity metric, and wherein determining that the at least two labels are related further comprises:

acquiring textual information associated with the set of content items;

determining, based on the textual information, a set of topic distributions for the set of content items, the set of topic distributions including a first topic distribution for a first content item in the set of content items and a second topic distribution for a second content item in the set of content items;

associating the first topic distribution with a first label out of the at least two labels and the second topic distribution with a second label out of the at least two labels, the first label being descriptive of the first content item and the second label being descriptive of the second content item;

calculating the topic similarity metric based on comparing the first topic distribution and the second topic distribution; and

determining that the topic similarity metric at least meets a specified topic similarity threshold.

20. The non-transitory computer-readable storage medium of claim 16, wherein the set of labels includes a set of hashtags.