Personalization engine for classifying unstructured documents

• US 8,214,346 B2
• Filed: 01/30/2009
• Issued: 07/03/2012
• Est. Priority Date: 06/27/2008
• Status: Active Grant

• Current Assignee: CBS Interactive Inc. (ViacomCBS Incorporated)
• Original Assignee: CBS Interactive Inc. (ViacomCBS Incorporated)
• Inventors: Pradhan, Tushar, Osborne, Thomas, Potter, John
• Primary Examiner(s): Vo, Tim T
• Assistant Examiner(s): TRAN, ANHTAI V
• Application Number:
  US12/362,840
• Publication Number:
  US 20090327243A1
• Time in Patent Office: 1,250 Days
• Field of Search: None
• US Class Current: 707/706
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking

Abstract

Unstructured electronic documents are classified for profiling and targeting users for additional relevant content. Behavioral data is gathered from user activity, and user documents and actions are categorized. Profile information is combined with collaborative and editorial data to provide users with credible information regarding products. Author-generated document classification information is analyzed and assigned a first taxonomic noun to characterize the document. User-generated tags characterizing a portion of the document are assigned a second taxonomic noun. Search terms that resulted in the user accessing the document are identified and assigned a third taxonomic noun. Attributes related to how the document was accessed are evaluated and assigned a fourth taxonomic noun. The document is processed using pattern rules to extract a fifth taxonomic noun. The taxonomic nouns are aggregated to determine term vectors representing the document, and the document is categorized using the term vectors, the taxonomic nouns, or the author-generated classification.

Claims

1. A computer-implemented method for classifying an electronic document, the method comprising:

analyzing, with a computing device, author-generated classification information regarding a document and assigning a set of first taxonomic nouns to characterize the document based upon the author-generated classification information;

examining, with a computing device, a user-generated tag from a client computer characterizing a portion of the document and assigning a set of second taxonomic nouns to characterize the document based upon the user-generated tag characterization;

identifying, with a computing device, a method of access through which the document has been accessed from a content provider and assigning a set of third taxonomic nouns to characterize the document based upon the method of access;

evaluating, with a computing device, attributes related to the method of access and assigning a set of fourth taxonomic nouns to characterize the document based upon the attributes related to the method of access;

processing, with a computing device, the document to extract a set of fifth taxonomic nouns to characterize the document based upon a predetermined pattern rule;

aggregating, with a computing device, the taxonomic nouns to determine at least one term vector that represents the document; and

categorizing, with a computing device, the document based upon the taxonomic nouns, the author-generated classification information, and at least one of the term vectors.

2. The computer-implemented method for classifying an electronic document of claim 1, wherein the author-generated information includes an author classification of the document characterizing a portion of the document.

3. The computer-implemented method for classifying an electronic document of claim 1, wherein the author-generated information includes metadata assigned to the document.

4. The computer-implemented method for classifying an electronic document of claim 1, wherein the attributes related to the method of access include at least one of a network type used to access the document, a Web site from which the document was accessed, an electronic correspondence to which the document was associated, a time of day the document was accessed, a referrer who directed a user to the document, and a document category used by a repository storing the document to describe the document.

5. The computer-implemented method for classifying an electronic document of claim 1, wherein the attributes related to the method of access include a click rate indicating a level of a user'"'"'s relative interest in the document in relation to other documents available to the user.

6. The computer-implemented method for classifying an electronic document of claim 1, wherein processing the document includes weighting the taxonomic nouns using a predetermined weighting scheme.

7. The computer-implemented method for classifying an electronic document of claim 1, wherein determining the at least one term vector includes building a list and a count of taxonomic nouns.

8. The computer-implemented method for classifying an electronic document of claim 1, wherein categorizing the electronic document based upon the at least one term vector includes processing the at least one term vector to derive a topic set.

9. The computer-implemented method for classifying an electronic document of claim 1, wherein processing the document to extract a set of fifth taxonomic nouns includes:

electronically reading the text of the document;

breaking the electronically read text into a series of words and punctuation characters;

comparing sequences of consecutive words with known terms read from an electronic dictionary database to determine if a match exists;

merging the matching consecutive words into a single multi-word term if a match exists and, if a match does not exist, assigning a single word term;

processing the multi-word term against a set of pattern rules with a pattern rules server to identify a part of speech assignment and a term type assignment; and

combining the terms and the part of speech assignment and the term type assignment into a structured term vector.

10. The computer-implemented method for classifying an electronic document of claim 7, wherein determining the at least one term vector includes applying a dictionary of taxonomic nouns with a dictionary building server.

11. The computer-implemented method for classifying an electronic document of claim 8, wherein categorizing the electronic document based upon the at least one term vector further includes executing a self search to identify additional taxonomic nouns to create an additional topic set.

12. The computer-implemented method for classifying an electronic document of claim 8, wherein categorizing the electronic document based upon the at least one term vector further includes executing a co-occurrence analysis to identify and cluster related taxonomic nouns to create a taxonomic noun cluster.

13. A system for classifying an electronic document, the system comprising:

a computing device configured for analyzing author-generated classification information regarding a document and assigning a set of first taxonomic nouns to characterize the document based upon the author-generated classification information;

a computing device configured for examining a user-generated tag from a client computer characterizing a portion of the document and assigning a set of second taxonomic nouns to characterize the document based upon the user-generated tag characterization;

a computing device configured for identifying a method of access through which the document has been accessed and assigning a set of third taxonomic nouns to characterize the document based upon the method of access;

a computing device configured for evaluating attributes related to the method of access and assigning a set of fourth taxonomic nouns to characterize the document based upon the attributes related to the method of access;

a computing device configured for processing the document to extract a set of fifth taxonomic nouns to characterize the document based upon a predetermined pattern rule;

a computing device configured for aggregating the taxonomic nouns to determine term vectors that represent the document; and

a computing device configured for categorizing the document based upon the taxonomic nouns, the author-generated classification information, and at least one of the term vectors.

14. The system for classifying an electronic document of claim 13, wherein the author-generated information includes an author classification of the document characterizing a portion of the document.

15. The system for classifying an electronic document of claim 13, wherein the author-generated information includes metadata assigned to the document.

16. The system for classifying an electronic document of claim 13, wherein the attributes related to the method of access include at least one of a network type used to access the document, a Web site from which the document was accessed, an electronic correspondence to which the document was associated, a time of day the document was accessed, a referrer who directed a user to the document, and a document category used by a repository storing the document to describe the document.

17. The system for classifying an electronic document of claim 13, wherein the attributes related to the method of access include a click rate indicating a level of a user'"'"'s relative interest in the document in relation to other documents available to the user.

18. The system for classifying an electronic document of claim 13, wherein processing the document includes weighting the taxonomic nouns using a predetermined weighting scheme.

19. The system for classifying an electronic document of claim 13, wherein determining the at least one term vector includes building a list and a count of taxonomic nouns.

20. The system for classifying an electronic document of claim 13, wherein categorizing the electronic document based upon the at least one term vector includes processing the at least one term vector to derive a topic set.

21. The system for classifying an electronic document of claim 13, wherein processing the document to extract a set of fifth taxonomic nouns includes:

electronically reading the text of the document;

breaking the electronically read text into a series of words and punctuation characters;

comparing sequences of consecutive words with known terms read from an electronic dictionary database to determine if a match exists;

merging the matching consecutive words into a single multi-word term if a match exists and, if a match does not exist, assigning a single word term;

processing the multi-word term against a set of pattern rules with a pattern rules server to identify a part of speech assignment and a term type assignment; and

combining the terms and the part of speech assignment and the term type assignment into a structured term vector.

22. The system for classifying an electronic document of claim 13, wherein a single computing device is configured to carry out the functions of two or more of the computing devices.

23. The system for classifying an electronic document of claim 19, wherein determining the at least one term vector includes applying a dictionary of taxonomic nouns with a dictionary building server.

24. The system for classifying an electronic document of claim 20, wherein categorizing the electronic document based upon the at least one term vector further includes executing a self search to identify additional taxonomic nouns to create an additional topic set.

25. The system for classifying an electronic document of claim 20, wherein categorizing the electronic document based upon the at least one term vector further includes executing a co-occurrence analysis to identify and cluster related taxonomic nouns to create a taxonomic noun cluster.

26. A computer program product with instructions recorded on a non-transitory computer readable storage medium, which, when executed by a processor, cause the processor to carry out a method for classifying an electronic document, the computer program product comprising:

instructions for analyzing, with a computing device, author-generated classification information regarding a document and assigning a set of first taxonomic nouns to characterize the document based upon the author-generated classification information;

instructions for examining, with a computing device, a user-generated tag from a client computer characterizing a portion of the document and assigning a set of second taxonomic nouns to characterize the document based upon the user-generated tag characterization;

instructions for identifying, with a computing device, a method of access through which the document has been accessed from a content provider and assigning at set of third taxonomic nouns to characterize the document based upon the method of access;

instructions for evaluating, with a computing device, attributes related to the method of access and assigning a set of fourth taxonomic nouns to characterize the document based upon the attributes related to the method of access;

instructions for processing, with a computing device, the document to extract a set of fifth taxonomic nouns to characterize the document based upon a predetermined pattern rule;

instructions for aggregating, with a computing device, the taxonomic nouns to determine at least one term vector that represents the document; and

instructions for categorizing, with a computing device, the document based upon the taxonomic nouns, the author-generated classification information, and at least one of the term vectors.

27. The computer program product of claim 26, wherein the author-generated information includes an author classification of the document characterizing a portion of the document.

28. The computer program product of claim 26, wherein the author-generated information includes metadata assigned to the document.

29. The computer program product of claim 26, wherein the attributes related to the method of access include at least one of a network type used to access the document, a Web site from which the document was accessed, an electronic correspondence to which the document was associated, a time of day the document was accessed, a referrer who directed a user to the document, and a document category used by a repository storing the document to describe the document.

30. The computer program product of claim 26, wherein the attributes related to the method of access include a click rate indicating a level of a user'"'"'s relative interest in the document in relation to other documents available to a user.

31. The computer program product of claim 26, wherein processing the document includes weighting the taxonomic nouns using a predetermined weighting scheme.

32. The computer program product of claim 26, wherein determining the at least one term vector includes building a list and a count of taxonomic nouns.

33. The computer program product of claim 26, wherein categorizing the electronic document based upon the at least one term vector includes processing the at least one term vector to derive a topic set.

34. The computer program product of claim 26, wherein processing the document to extract a set of fifth taxonomic nouns includes:

electronically reading the text of the document;

breaking the electronically read text into a series of words and punctuation characters;

comparing sequences of consecutive words with known terms read from an electronic dictionary database to determine if a match exists;

merging the matching consecutive words into a single multi-word term if a match exists and, if a match does not exist, assigning a single word term;

processing the multi-word term against a set of pattern rules with a pattern rules server to identify a part of speech assignment and a term type assignment; and

combining the terms and the part of speech assignment and the term type assignment into a structured term vector.

35. The computer program product of claim 32, wherein determining the at least one term vector includes applying a dictionary of taxonomic nouns with a dictionary building server.

36. The computer program product of claim 33, wherein categorizing the electronic document based upon the at least one term vector further includes executing a self search to identify additional taxonomic nouns to create an additional topic set.

37. The computer program product of claim 33, wherein categorizing the electronic document based upon the at least one term vector further includes executing a co-occurrence analysis to identify and cluster related taxonomic nouns to create a taxonomic noun cluster.