Pseudo-anchor text extraction

• US 8,073,838 B2
• Filed: 01/29/2010
• Issued: 12/06/2011
• Est. Priority Date: 03/02/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Shi, Shuming, Wen, Ji-Rong, Zhu, Mingjie, Xing, Fei, Nie, Zaiqing
• Primary Examiner(s): Ehichioya, Fred I
• Application Number:
  US12/697,056
• Publication Number:
  US 20100145956A1
• Time in Patent Office: 676 Days
• Field of Search: 707/609, 707/637, 707/705, 707/713, 707/769, 707/802, 707/821, 709/223, 709/224, 709/231, 719/316, 719/317, 719/320
• US Class Current: 707/713
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing

Abstract

A search method uses pseudo-anchor text associated with search objects to improve search performance. The pseudo-anchor text may be extracted in combination with an identifier of the search objects (such as a pseudo-URL) from a digital corpus such as a collection of documents. Pseudo-anchor texts for each object are preferably extracted from candidate anchor blocks using a machine learning based approach. The pseudo-anchor texts are made available for searching and used to help rank the objects in a search result to improve search performance. The method may be used in vertical search of objects such as published articles, products and images that lack explicit URLs and anchor text information.

Claims

1. A method implemented by a computer, the method comprising:

identifying an object in content;

identifying a first block of information associated with the object and a second block of information associated with the object;

extracting the first block of information associated with the object and the second block of information associated with the object;

aggregating the first block of information associated with the object and the second block of information associated with the object to obtain aggregated information associated with the object;

constructing, by the computer, a pseudo uniform resource locator (pseudo-URL) of the object based at least on the information being extracted and aggregated;

associating the pseudo-URL with the object;

storing for repeated use, the pseudo-URL being associated with the object and the information being extracted; and

providing, via the pseudo-URL, the information extracted for search.

2. A computer-implemented method as recited in claim 1, wherein the object lacks an associated uniform resource locator (URL).

3. A computer-implemented method as recited in claim 1, further comprising extracting pseudo-anchor text from the aggregated information associated with the object, wherein the pseudo-anchor text comprises the information extracted that is provided for search.

4. A computer-implemented method as recited in claim 1, wherein the content is one of a plurality of types of content, the plurality of types of content comprising a document, a compilation, an image, or a video.

5. A computer-implemented method as recited in claim 1, further comprising ranking search results based at least on the information extracted.

6. A computer-implemented method as recited in claim 1, further comprising parsing the content to construct a link associated with the pseudo-URL.

7. A computer-readable media having computer-readable instructions encoded thereon, the computer-readable instructions upon execution configuring a computer to perform the method recited in claim 1.

8. A system comprising:

a processor;

a memory coupled to the processor, the memory storing a plurality of modules, at least one of the plurality of modules upon execution by the processor performing pseudo-anchor text extraction comprising;

identifying an object in content;

identifying a first candidate anchor block associated with the object and a second candidate anchor block associated with the object;

extracting pseudo-anchor text from the first candidate anchor block and the second candidate anchor block, wherein the pseudo-anchor text being extracted includes text including a possible description of the object;

accumulating the pseudo-anchor text extracted from the first candidate anchor block and the pseudo-anchor text extracted from the second candidate anchor block to obtain pseudo-anchor text associated with the object;

constructing a pseudo uniform resource locator (pseudo-URL) of the object based at least on the pseudo-anchor text associated with the object;

associating the pseudo-URL with the object;

storing for repeated use, the pseudo-URL being associated with the object and the pseudo-anchor text associated with the object; and

providing for search, via the pseudo-URL, the pseudo-anchor text associated with the object.

9. A system as recited in claim 8, wherein the at least one of the plurality of modules includes a pseudo uniform resource locator (pseudo-URL) extraction module that upon execution by the processor causes extraction of the pseudo-URL.

10. A system as recited in claim 8, wherein the at least one of the plurality of modules includes a candidate anchor block extraction module that upon execution by the processor causes the pseudo-anchor text from the first candidate anchor block and the pseudo-anchor text from the second candidate anchor block, to be extracted.

11. A system as recited in claim 10, wherein the at least one of the plurality of modules further includes a candidate anchor block accumulation module that upon execution by the processor causes the pseudo-anchor text extracted from the first candidate anchor block and the pseudo-anchor text extracted from the second candidate anchor block to be accumulated.

12. A system as recited in claim 8, wherein the at least one of the plurality of modules includes a pseudo-anchor extraction module to identify and extract the pseudo-anchor text associated with the object for use in search.

13. A system as recited in claim 8, further comprising a digital corpus available for pseudo-anchor text extraction.

14. A system as recited in claim 13, wherein the digital corpus comprises one or more of a plurality of types of content, the plurality of types of content comprising a document, a compilation, an image, or a video.

15. A computer-readable media having computer-readable instructions encoded thereon, the computer-readable instructions upon execution configuring a search engine to perform operations comprising:

identifying an object in content;

identifying a first candidate anchor block associated with the object and a second candidate anchor block associated with the object;

extracting the first candidate anchor block associated with the object and the second candidate anchor block associated with the object;

aggregating the first candidate anchor block associated with the object and the second candidate anchor block associated with the object to obtain pseudo-anchor text associated with the object;

extracting the pseudo-anchor text from the candidate anchor blocks that were aggregated;

constructing a pseudo uniform resource locator (pseudo-URL) of the object based at least on the extracted pseudo-anchor text;

associating the pseudo-URL with the object;

storing for repeated use, the pseudo-URL being associated with the object and the extracted pseudo-anchor text; and

providing, via the pseudo-URL, the pseudo-anchor text extracted for search.

16. A computer-readable media as recited in claim 15, wherein the object lacks an associated uniform resource locator (URL).

17. A computer-readable media as recited in claim 15, wherein the content comprises one or more of a plurality of types of content, the plurality of types of content comprising a document, a compilation, an image, or a video.

18. A computer-readable media as recited in claim 15, the operations further comprising:

identifying pseudo-URLs for a plurality of objects;

parsing the content to construct a pseudo-link associated with the pseudo-URL.

19. A computer-readable media as recited in claim 15, the operations further comprising ranking search results based at least on the pseudo-anchor text.

20. A computer-readable media as recited in claim 15, the operations further comprising automatically assigning each term in each candidate anchor block a degree of belonging to pseudo-anchor text.