Spam detection for user-generated multimedia items based on concept clustering

• US 8,745,056 B1
• Filed: 03/31/2008
• Issued: 06/03/2014
• Est. Priority Date: 03/31/2008
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Kacholia, Varun, Garg, Ashutosh, Stoutamire, David
• Primary Examiner(s): Hoang, Son T
• Application Number:
  US12/059,135
• Time in Patent Office: 2,255 Days
• Field of Search: 707/735, 707/737, 707/758
• US Class Current: 707/737
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/7867 using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings

Abstract

A system, a method, and various software tools enable a video hosting website to automatically identify posted video items that contain spam in the metadata associated with a respective video item. A spam detection tool for user-generated video items based on concept clustering is provided that facilitates the detection of spam in the metadata associated with a video item.

Claims

1. A computer-implemented method for processing a video, comprising:

storing the video and associated metadata in a multimedia database of a hosting website, wherein the metadata includes a plurality of tokens that are provided from a user who uploaded the video to the hosting website;

determining, for each of the plurality of tokens, a concept associated with the token;

determining a number of concepts related to content of the video and a number of distinct concepts unrelated to the content of the video from concepts determined from the plurality of tokens; and

responsive to the number of distinct concepts unrelated to the content of the video exceeding a threshold, marking the video as spam.

2. The computer-implemented method of claim 1, further comprising removing the video from the multimedia database when the video is marked as spam.

3. The computer-implemented method of claim 1, further comprising refraining from indexing the video in the multimedia database when the video is marked as spam.

4. The computer-implemented method of claim 1, wherein the threshold is based on a distribution of a number of concepts related to the content of the video that are not indicative of spam.

5. The computer-implemented method of claim 1, wherein the threshold is based on a mean number of concepts unrelated to the content of the video.

6. A computer-implemented method for processing a video, comprising:

storing the video and associated metadata in a multimedia database of a hosting website, wherein the metadata includes a plurality of tokens that are provided from a user who uploaded the video to the hosting website;

determining, for each of the plurality of tokens, a concept associated with the token;

determining a number of concepts related to content of the video and a number of distinct concepts unrelated to the content of the video from concepts determined from the plurality of tokens; and

responsive to determining at least one combination of distinct concepts unrelated to the content of the video, marking the video as spam.

7. The computer-implemented method of claim 6, further comprising removing the video from the multimedia database when the video is marked as spam.

8. The computer-implemented method of claim 6, further comprising refraining from indexing the video when the video is marked as spam.

9. A system for processing a video, the system comprising:

a computer processor; and

a computer-readable storage medium storing executable code, the code when executed by the processor performs steps comprising;

storing the video and associated metadata in a multimedia database of the system, wherein the metadata includes a plurality of tokens that are provided from a user who uploaded the video to the system;

determining, for each of the plurality of tokens, a concept associated with the token;

determining a number of concepts related to content of the video and a number of distinct concepts unrelated to the content of the video from concepts determined from the plurality of tokens; and

marking the video as spam responsive to the number of distinct concepts unrelated to the content of the video exceeding a threshold.

10. The system of claim 9, wherein the code when executed by the processor performs a further step of removing the video from the multimedia database when the video is marked as spam.

11. The system of claim 9, wherein the code when executed by the processor performs further steps comprising for refraining from indexing the video when the video is marked as spam.

12. A non-transitory computer-readable storage medium containing program code for processing a video, the program code for:

storing the video and associated metadata in a multimedia database of a hosting website, wherein the metadata includes a plurality of tokens that are provided from a user who uploaded the video to the hosting website;

determining, for each of the plurality of tokens, a concept associated with the token;

determining a number of concepts related to content of the video and a number of distinct concepts unrelated to the content of the video from concepts determined from the plurality of tokens; and

responsive to the number of distinct concepts unrelated to the content of the video exceeding a threshold, marking the video as spam.

13. The computer-readable storage medium of claim 12, further comprising removing the video from a multimedia database when the video is marked as spam.

14. The computer-readable storage medium of claim 12, further comprising program code for refraining from indexing the video when the video is marked as spam.