Associating social comments with individual assets used in a campaign

• US 10,339,559 B2
• Filed: 12/04/2014
• Issued: 07/02/2019
• Est. Priority Date: 12/04/2014
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Inc.
• Inventors: Duggal, Ashish, Dhawan, Anmol, Soni, Sachin, Jain, Anuj
• Primary Examiner(s): Sigmond, Bennett M
• Assistant Examiner(s): Anderson, Scott C
• Application Number:
  US14/560,791
• Publication Number:
  US 20160162930A1
• Time in Patent Office: 1,671 Days
• CPC Class Codes: G06Q 30/0245 Surveys
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Associating social comments with individual assets used in a campaign is described. In one or more embodiments, a campaign that includes one or more assets (e.g., images or videos of products) is published to one or more social networks. Comments (e.g., user comments, user shares, or other textual feedback) to the campaign on the one or more social networks are collected and analyzed to attribute each comment to an individual asset of the campaign. Social metadata, such as a social mention count and a social sentiment score, is generated based on the comments to enhance metadata of the individual asset.

Claims

1. A system comprising:

at least one processor; and

a computer-readable storage medium storing instructions that are executable by the at least one processor to perform operations comprising;

collecting a comment to a campaign, the campaign including a plurality of different assets and including embedded metadata associated with each of the plurality of different assets;

analyzing the comment to extract a keyword from the comment;

comparing the extracted keyword to the embedded metadata associated with each of the plurality of different assets;

attributing the comment to one of the plurality of different assets based on the comparison;

generating social metadata for the one of the plurality of different assets, based on the content of the comment; and

embedding the generated social metadata in the one of the plurality of different assets and outputting the asset with the embedded social metadata to an asset database.

2. The system as recited in claim 1, the operations further comprising:

determining a social sentiment score of the comment; and

determining a total social sentiment score for the one of the plurality of different assets based on the social sentiment score of the comment, the generated social metadata describing the total social sentiment score.

3. The system as recited in claim 2, the operations further comprising:

determining a social mention count for the one of the plurality of different assets based on a number of comments submitted to the campaign; and

determining an average social sentiment score for the one of the plurality of different assets based on a ratio between the social mention count and the total sentiment score.

4. The system as recited in claim 3, wherein the social metadata for the one of the plurality of different assets includes the average social sentiment score for the one of the plurality of different assets.

5. The system as recited in claim 4, wherein the asset database is searchable based on average sentiment score or total social sentiment score.

6. The system as recited in claim 2, further comprising determining a performance of the one of the plurality of different assets based on at least one of the social mention count and the social sentiment score.

7. The system as recited in claim 1, wherein the comparing further comprises:

determining a cosine similarity between the extracted keyword and keywords associated with the plurality of different assets; and

,

attributing the comment to the one of the plurality of different assets is urther in response to the cosine similarity satisfying a threshold.

8. A computer-implemented method comprising:

storing, by a computing device and for each of a plurality of different assets in an asset database, an asset identifier and metadata associated with the asset;

creating, by the computing device, a campaign that includes at least two of the plurality of different assets;

embedding, by the computing device, metadata associated with the at least two different assets into the campaign;

publishing, by the computing device, the campaign with the embedded metadata on at least one social network;

collecting, by the computing device, a comment to the campaign submitted via the at least one social network;

analyzing, by the computing device, the collected comment to identify at least one keyword in the comment;

attributing, by the computing device, the collected comment to one of the at least two different assets included in the campaign based on a detected similarity between the at least one identified keyword in the comment and asset metadata embedded in the campaign;

generating, by the computing device, social metadata for the asset to which the comment is attributed, based on content of the comment; and

embedding, by the computing device, the generated social metadata in the metadata of the attributed asset in the asset database.

9. The computer-implemented method as recited in claim 8, wherein the social metadata describes a social mention count of the asset to which the comment is attributed.

10. The computer-implemented method as recited in claim 8, wherein analyzing the collected comment is performed using a natural language processing technique to extract the keyword.

11. The computer-implemented method as recited in claim 8, wherein attributing the collected comment to one of the at least two different assets is further based on:

determining a cosine similarity between the at least one keyword in the comment and keywords associated with the plurality of different assets; and

determining that the cosine similarity satisfies a threshold.

12. The computer-implemented method as recited in claim 8, wherein the social metadata describes a social sentiment score of the collected comment on a scale of negative to positive sentiment.

13. The computer-implemented method as recited in claim 8, wherein the asset database is searchable based on the social metadata.

14. A computer-implemented method comprising:

storing, by a computing device, for each asset in an asset database, an asset identifier and metadata associated with the asset identifier;

collecting, by the computing device, a comment to a campaign, the campaign including a plurality of different assets and including embedded metadata associated with the plurality of different assets;

analyzing, by the computing device, the comment using natural language processing to extract a keyword from the comment;

comparing, by the computing device, the extracted keyword to the embedded metadata associated with the plurality of different assets in the campaign;

attributing, by the computing device, the comment to one of the plurality of different assets based on the comparison;

determining, by the computing device, a social mention count for the one of the plurality of different assets, the social mention count indicating a total number of comments to the campaign that mention the one of the plurality of different assets;

determining, by the computing device, a social sentiment score of the comment;

generating, by the computing device, social metadata for the one of the plurality of different assets that describes the social mention count and the social sentiment score; and

embedding the generated social metadata into the asset database metadata associated with the one of the plurality of different assets.

15. The computer-implemented method as recited in claim 14, wherein the comparing further comprises:

determining, by the computing device, a cosine similarity between the extracted keyword and keywords associated with the plurality of different assets, wherein attributing the comment to the one of the plurality of different assets is further based on the cosine similarity satisfying a threshold.

16. The computer-implemented method as recited in claim 14, further comprising determining, by the computing device, a performance of the one of the plurality of different assets based on the social mention count and the social sentiment score.

17. The computer-implemented method as recited in claim 14, further comprising determining, by the computing device, an average social sentiment score for the one of the plurality of different assets based on a ratio of the total social sentiment score to the social mention count.

18. The computer-implemented method as recited in claim 17, wherein the asset database is searchable based on the average social sentiment score.

19. The computer-implemented method as recited in claim 14, wherein the social sentiment score describes a sentiment towards the comment on a scale of negative sentiment to positive sentiment.

20. The computer-implemented method as recited in claim 14, further comprising generating, by the computing device, a second campaign that includes one of the plurality of different assets based on the social metadata.