Method and apparatus for detecting spam user created content

• US 8,095,547 B2
• Filed: 12/28/2007
• Issued: 01/10/2012
• Est. Priority Date: 09/27/2007
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Jeong, Ho Wook, Kang, Jae Ho, Kim, Hyung Deok
• Primary Examiner(s): Pham, Khanh
• Assistant Examiner(s): CHEEMA, AZAM M
• Application Number:
  US11/966,388
• Publication Number:
  US 20090089279A1
• Time in Patent Office: 1,474 Days
• Field of Search: None
• US Class Current: 707/752
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: H04L 51/212 using filtering or selective blocking

Abstract

A method includes receiving user-created content, as well as at least one of a service identifier (ID) and a content category ID of the user-created content. At least one of a plurality of key information databases is selected based on at least one of the received service ID and the received content category ID. Second key information is extracted from the received user-created content. The selected key information database is searched to retrieve first key information related to the second key information. The user-created content may be classified as spam content based on the extracted second key information and/or the retrieved first key information. The user-created content is conditionally stored in a network accessible data store, available to users of the user-created content hosting site, based upon the classification.

Claims

1. A method for processing spam contents, comprising the steps of:

maintaining a plurality of key information databases;

receiving user-created content and at least one of a service identifier (ID) and a content category ID of said user-created content from one or more users of a user-created content hosting site;

selecting at least one of the plurality of key information databases based on at least one of the received service ID and the received content category ID;

extracting second key information from the received user-created content;

searching the selected key information database to retrieve first key information related to the second key information;

classifying the user-created content as spam content based on the extracted second key information and/or the retrieved first key information related to the second key information; and

conditionally storing the user-created content in a network accessible data store available to users of the user-created content hosting site based on classifying the user-created content as spam or non-spam content

wherein,

the second key information is different from the service ID, and

the second key information is different from the content category ID.

2. The method of claim 1, wherein said first and second key information comprise at least one of predetermined type(s) of data, word(s) and phrase(s) in said contents, and wherein said data comprises at least one of a user ID, a universal resource locator, a site address, an account number or a telephone number.

3. The method of claim 2, wherein said maintaining comprises maintaining a spam information database, and wherein said classifying comprises classifying said user-created content as spam content based on whether the spam information database includes the extracted second key information.

4. The method of claim 3, wherein said maintaining a spam information database comprises:

preparing a series of user-created contents, each of said contents classified as spam or non-spam content;

obtaining, from the series of user-created contents, third key information and the frequency the third key information is found in the contents classified as spam content; and

recording the third key information in the spam information database based on the obtained frequency.

5. The method of claim 1, wherein said classifying comprises classifying said user-created content as spam content using a document classification algorithm.

6. The method of claim 5, wherein said classifying comprises:

obtaining spam probability data for the extracted second key information from the key information database; and

classifying the user-created content as spam content based on the obtained spam probability data.

7. The method of claim 6, wherein said updating comprises updating the spam probability data for the extracted second key information in accordance with the extracted second key information.

8. The method of claim 1, further comprising:

updating the key information database based on the extracted second key information.

9. The method of claim 8, wherein said updating comprises registering the extracted second key information as the first key information in the key information database.

10. The method of claim 8, wherein said updating comprises updating the frequency the extracted second key information is found in spam contents in accordance with the extracted second key information.

11. The method of claim 1, further comprising:

determining whether the extracted second key information corresponds to predefined restricted information; and

if the extracted key second information corresponds to the predefined restricted information, removing the extracted second key information or replacing the extracted second key information with predefined different information.

12. Logic encoded in one or more tangible media for execution and when executed operable to cause the one or more processors to:

maintain a plurality of key information databases;

receive user-created content at least one of a service identifier (ID) and a content category ID of said user-created content from one or more users of a user-created content hosting site;

select at least one of the plurality of key information databases based on at least one of the received service ID and the received content category ID;

extract second key information from the received user-created content;

search the selected key information database to retrieve first key information related to the second key information;

classify the user-created content as spam content based on the extracted second key information and/or the retrieved first key information related to the second key information; and

conditionally store the user-created content in a network accessible data store available to users of the user-created content hosting site based on classifying the user-created content as spam or non-spam content

wherein,

the second key information is different from the service ID, and

the second key information is different from the content category ID.

13. An apparatus for processing spam contents, said apparatus comprising:

a non-transitory computer-readable medium configured to include a plurality of key information databases;

a communication part configured to receive user-created content and at least one of a service ID and a content category ID from one or more of users of a user-created content hosting site; and

a control part configured to select one of the plurality of key information databases based on at least one of the service ID and the content category ID, extract second key information from the received user-created content, search the selected key information database to retrieve first key information related to the extracted second key information, classify the received user-created content as spam content based on the extracted second key information and/or the retrieved first key information related to the first key information, and conditionally store the user-created content in a network accessible data store available to users of the user-created content hosting site based on classifying the user-created content as spam or non-spam content

wherein,

the second key information is different from the service ID, and

the second key information is different from the content category ID.

14. The apparatus of claim 13, wherein said first and second key information comprise at least one of predetermined types of data, a word and a phrase of said contents, said data comprising at least one of a user ID, a URL (Universal Resource Locator), a site address, an account number and a telephone number.

15. The apparatus of claim 13, wherein said control part comprises:

a key information management part configured to prepare a spam information database by classifying at least some of key information of the first information as spam key information; and

a spam information matching part configured to classify said contents as spam contents based on whether the spam information database includes the extracted second key information.

16. The apparatus of claim 15, wherein said key information management part prepares a series of user-created contents, each of said user-created content classified as spam or non-spam content;

obtains, from the series of user-created contents, third key information and the frequency the third key information is found in the user-created content classified as spam content; and

records the third key information in the spam information database based on the obtained frequency.

17. The apparatus of claim 13, wherein said control part comprises a document classifying part configured to classify said user-created content as spam content using a document classification algorithm.

18. The apparatus of claim 17, wherein said document classifying part obtains spam probability data for the extracted second key information from the key information database and classifies the user-created content as spam content based on the obtained spam probability data.

19. The apparatus of claim 13, wherein said control part comprises a key information management part configured to update the key information database based on the extracted second key information.

20. The apparatus of claim 19, wherein said key information management part registers the extracted second key information as the first key information in the key information database.

21. The apparatus of claim 19, wherein said key information management part updates the frequency the extracted second key information is found in spam content in accordance with the extracted second key information.

22. The apparatus of claim 19, wherein said key information management part updates the spam probability data for the extracted second key information in accordance with the extracted second key information.

23. The apparatus of claim 13, wherein said control part further comprises:

a contents management part configured to determine whether the extracted second key information corresponds to predefined restricted information and, if the extracted second key information corresponds to the predefined restricted information, removes the extracted second key information or replaces the extracted second key information with predefined different information.

24. A method for processing spam contents, comprising the steps of:

maintaining a plurality of key information databases;

receiving user-created content and at least one of a service ID and a content category ID of the user-created content from one or more users of a user-created content hosting site;

selecting one of the plurality of key information databases based on at least one of the service ID and the content category ID;

extracting second key information from the received user-created content;

searching the selected key information database to retrieve first key information matching the second key information;

if a match is found, classifying the user-created content as spam content; and

conditionally storing the user-created content in a network accessible data store available to users of the user-created content hosting site based on classifying the user-created content as spam or non-spam content

wherein,

the second key information is different from the service ID, and

the second key information is different from the content category ID.

25. An apparatus for processing spam contents, said apparatus comprising:

a non-transitory computer-readable medium configured to include a key information database;

a communication part configured to receive user-created content and at least one of a service ID and a content category ID of the user-created content from one or more users of a user-created content hosting site; and

a control part configured to select one of the plurality of key information databases based on at least one of the service ID and the content category ID, extract second key information from the received user-created content, search the selected key information database to retrieve first key information matching the second key information, if a match is found, classify the user-created content as spam content and conditionally store the user-created content in a network accessible data store available to users of the user-created content hosting site based on classifying the user-created content as spam or non-spam content

wherein,

the second key information is different from the service ID, and

the second key information is different from the content category ID.