Method and system for characterizing web content

• US 9,213,767 B2
• Filed: 08/10/2009
• Issued: 12/15/2015
• Est. Priority Date: 08/10/2009
• Status: Active Grant

• Current Assignee: Micro Focus LLC (Open Text Corporation)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Scholz, Martin B., Rajaram, Shyam Sundar, Lukose, Rajan
• Primary Examiner(s): Hoang, Son T
• Application Number:
  US12/538,405
• Publication Number:
  US 20110035378A1
• Time in Patent Office: 2,318 Days
• Field of Search: 707/E17.044, 707/E17.108
• US Class Current: 1/1
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

An exemplary embodiment of the present invention provides a method of processing Web activity data. The method includes obtaining a database of Website organizational data. The method also includes generating a data structure from the database of Website organizational data comprising an Item identifier and a Website category corresponding to the item identifier. The method also includes generating a reduced-rank classification structure from the data structure, the reduced-rank classification structure including a category grouping corresponding to one or more of the Website categories.

Claims

1. A method of processing Web activity data, comprising:

obtaining, by a computer processor, a database of Website organizational data at a networked computer from a plurality of Websites, the Website organizational data comprising a plurality of Website-specific category names and corresponding item identifiers, wherein each of the plurality of Website-specific category names is associated with one or more of the corresponding item identifiers and each item identifier uniquely identifying a particular item within the plurality of Websites;

generating, by the computer processor, a data structure in the networked computer from the database of Website organizational data, the data structure comprising a matrix including a plurality of entries, wherein

each entry of the matrix comprising one of the corresponding item identifiers on a first axis of the matrix and one of the plurality of Website-specific category names on a second axis of the matrix, and

each entry in the matrix indicates that the one of the corresponding item identifiers is organized under the one of the plurality of Website-specific category names;

generating, by the computer processor, a reduced-rank classification structure in the networked computer from the data structure via a matrix decomposition including at least a singular-value decomposition of the data structure, wherein the reduced-rank classification structure combines the plurality of the Website-specific category names from multiple Websites of the plurality Websites into a single combined category-name grouping used to categorize content of the multiple Websites; and

storing, by the computer processor, the reduced-rank classification structure at the networked computer.

2. The method of claim 1, wherein each of the plurality of the Website-specific category names comprises a unit of a Website'"'"'s taxonomy or a tag.

3. The method of claim 1, wherein the generating the data structure comprises filling the matrix depending on a level of certainty that a specific item identifier corresponds with a specific Website-specific category name.

4. The method of claim 1, further comprising generating an approximation of the Website organizational data based on the reduced-rank classification structure.

5. The method of claim 4, wherein the generating the approximation of the Website organizational data comprises performing a sparse regression of the reduced-rank classification structure.

6. The method of claim 1, wherein the reduced-rank classification structure is used to generate a user profile.

7. The method of claim 1, wherein the reduced-rank classification structure represents a set of vectors corresponding to positions within a latent vector space.

8. The method of claim 7, further comprising mapping a new Website'"'"'s organizational data to the latent space of the reduced-rank classification structure.

9. A computer system, comprising:

a computer processor to execute computer-readable instructions;

a storage device to store a database of Website organizational data received from a plurality of Websites, the Website organizational data comprising a plurality of Website-specific category names and corresponding item identifiers, wherein each of the plurality of Website-specific category names is associated with one or more of the corresponding item identifiers and each item identifier uniquely identifying a particular item within the plurality of Websites; and

a memory device that stores instructions, executed by the computer processor, to direct the computer processor to;

generate a data structure from the database of Website organizational data, the data structure comprising a matrix including a plurality of entries, wherein

each entry of the matrix comprising one of the corresponding item identifiers on a first axis of the matrix and one of the plurality of Website-specific category names on a second axis of the matrix, and

each entry in the matrix indicates that the one of the corresponding item identifiers is organized under the one of the plurality of Website-specific category names; and

generate a reduced-rank classification structure from the data structure via a matrix decomposition including at least a singular-value decomposition of the data structure, wherein the reduced-rank classification structure combines the plurality of the Website-specific category names from multiple Websites of the plurality Websites into a single combined category-name grouping used to categorize content of the multiple Websites; and

store the reduced-rank classification structure at the storage device.

10. The computer system of claim 9, wherein each of the plurality of the Website-specific category names comprises a unit of a Website'"'"'s taxonomy or a tag.

11. The computer system of claim 9, wherein the generating the data structure comprises filling the matrix depending on a level of certainty that a specific item identifier corresponds with a specific Website-specific category name.

12. The computer system of claim 9, further comprising generating an approximation of the Website organizational data by performing a sparse regression of the reduced-rank classification structure.

13. The computer system of claim 9, wherein the reduced-rank classification structure is used to generate a user profile.

14. The computer system of claim 9, wherein the reduced-rank classification structure represents a set of vectors corresponding to positions within a latent vector space.

15. The computer system of claim 14, further comprising mapping a new Website'"'"'s organizational data to the latent space of the reduced-rank classification structure.

16. A non-transitory computer-readable medium, comprising code configured to direct a computer processor to:

obtain a database of Website organizational data at a networked computer from a plurality of Websites, the Website organizational data comprising a plurality of Website-specific category names and corresponding item identifiers, wherein each of the plurality of Website-specific category names is associated with one or more of the corresponding item identifiers and each item identifier uniquely identifying a particular item within the plurality of Websites;

generate a data structure in the networked computer from the database of Website organizational data, the data structure comprising a matrix including a plurality of entries, wherein

each entry of the matrix comprising one of the corresponding item identifiers on a first axis of the matrix and one of the plurality of Website-specific category names on a second axis of the matrix, and

each entry in the matrix indicates that the one of the corresponding item identifiers is organized under the one of the plurality Website-specific category names; and

generate a reduced-rank classification structure in the networked computer from the data structure via a matrix decomposition including at least a singular-value decomposition of the data structure, wherein the reduced-rank classification structure combines the plurality of the Website specific category names from multiple Websites of the plurality Websites into a single combined category-name grouping used to categorize content of the multiple Websites; and

storing, by the computer processor, the reduced-rank classification structure at the networked computer.

17. The non-transitory computer-readable medium of claim 16, wherein the generating the data structure comprises filling the matrix depending on a level of certainty that a specific item identifier corresponds with a specific Website-specific category name.