Authoritative document identification

• US 8,650,197 B2
• Filed: 03/09/2012
• Issued: 02/11/2014
• Est. Priority Date: 12/30/2004
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Egnor, Daniel, Chaudry, Geeta
• Primary Examiner(s): Hwa, Shyue Jiunn
• Application Number:
  US13/417,026
• Publication Number:
  US 20120173544A1
• Time in Patent Office: 704 Days
• Field of Search: 707/741
• US Class Current: 707/748
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9537 Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

A system determines documents that are associated with a location, identifies a group of signals associated with each of the documents, and determines authoritativeness of the documents for the location based on the signals.

Claims

1. A computer implemented method for identifying an authoritative web page corresponding to a business, the method comprising:

identifying, using a processor associated with the computer, a plurality of candidate web pages that are associated with the business;

identifying, using the processor, a plurality of signals respectively associated with each of the plurality of candidate web pages;

determining, using the processor, an authoritative score for each of the plurality of candidate web pages based on the plurality of signals respectively associated with each of the plurality of candidate web pages,

the authoritative score for each web page being determined based on one or more of;

a number of outlinks, in one or more of the plurality of candidate web pages, that point to the web page,

a match between anchor text associated with the outlinks that point to the web page and a name of the business,

a match between a title of the web page and the name of the business,

a number of geographic locations identified in the web page, or

a match between a domain name associated with the web page and the name of the business,

determining the authoritative score for each web page of the plurality of candidate web pages including;

weighting and combining a respective plurality of scores associated with the respective plurality of signals for each web page to determine the authoritative score for each web page; and

identifying, using the processor, a particular web page of the plurality of candidate web pages as an authoritative web page for the business,

the particular web page having a highest authoritative score of the authoritative scores for the plurality of candidate web pages.

2. The method of claim 1, where identifying the plurality of candidate web pages includes:

analyzing web pages in a web page corpus to identify snippets of text that include information associated with the business, and

identifying web pages that include the snippets of text as candidate web pages.

3. The method of claim 2, where the information associated with the business includes at least one of:

at least a portion of an address associated with a location of the business,

at least a portion of a telephone number associated with the business, or

at least a portion of the name of the business.

4. The method of claim 2, where identifying the plurality of candidate web pages further includes:

identifying web pages that are linked to the web pages that include the snippets of text as candidate web pages.

5. The method of claim 2, where identifying the plurality of candidate web pages further includes:

identifying one or more geographic addresses, that are not associated with a location of the business, in the plurality of candidate web pages, and

identifying additional web pages in the corpus of web pages that are also associated with the one or more geographic addresses.

6. The method of claim 1, where identifying the plurality of candidate web pages includes:

removing one or more portions of an address of a particular web page included in the plurality of candidate web pages to identify one or more additional web pages as candidate web pages.

7. A system to identify an authoritative web page corresponding to a business, the system comprising:

one or more computers configured to;

identify a plurality of candidate web pages that are associated with the business;

identify a plurality of signals respectively associated with each of the plurality of candidate web pages;

determine an authoritative score for each of the plurality of candidate web pages based on the plurality of signals respectively associated with each of the plurality of candidate web pages,

the authoritative score for each web page being determined based on one or more of;

a number of outlinks in one or more of the plurality of candidate web pages that point to the web page,

a match between anchor text associated with the outlinks that point to the web page and a name of the business,

a match between a title of the web page and the name of the business,

a number of geographic locations identified in the web page, or

a match between a domain name associated with the web page and the name of the business,

when determining the authoritative score for each candidate web page of the plurality of candidate web pages, the one or more computers being to;

weight and combine a respective plurality of scores associated with the respective plurality of signals for each candidate web page to determine the authoritative score for each candidate web page; and

identify a particular web page of the plurality of candidate web pages as an authoritative web page for the business,

the particular web page having a highest authoritative score of the authoritative scores for the plurality of candidate web pages.

8. The system of claim 7, where the one or more computers, when identifying the plurality of candidate web pages, are further to:

analyze web pages in a web page corpus to identify snippets of text that include information associated with the business; and

identify web pages that include the snippets of text as candidate web pages.

9. The system of claim 8, where the information associated with the business includes at least one of:

at least a portion of an address associated with a particular location of the business,

at least a portion of a telephone number associated with the business, or

at least a portion of the name of the business.

10. The system of claim 8, where the one or more computers, when identifying the plurality of candidate web pages, are further to:

identify web pages that are linked to the web pages that include the snippets of text as candidate web pages.

11. The system of claim 10, where the one or more computers, when identifying the plurality of candidate web pages, are further to:

identify one or more geographic addresses that are not associated with a location of the business, in the plurality of candidate web pages, and

identify additional web pages in the corpus of web pages that are also associated with the one or more geographic addresses as candidate web pages.

12. The system of claim 7, where the one or more computers, when determining the authoritative score, are further to:

increase the authoritative score, for one of the plurality of candidate web pages when the number of geographic locations associated with the one of the plurality of candidate web pages is less than a threshold.

13. The system of claim 7, where, when identifying the plurality of candidate web pages, the one or more computers are configured to:

remove one or more portions of an address of a particular web page included in the plurality of candidate web pages to identify one or more additional web pages as candidate web pages.

14. A non-transitory computer-readable medium to store instructions, the instructions comprising:

one or more instructions which, when executed by a processor, cause the processor to identify a plurality of candidate web pages that are associated with a business;

one or more instructions which, when executed by the processor, cause the processor to identify a plurality of signals respectively associated with each of the plurality of candidate web pages;

one or more instructions which, when executed by the processor, cause the processor to determine an authoritative score for each of the plurality of candidate web pages based on the respective plurality of signals,

the authoritative score for each web page being determined based on one or more of;

a number of outlinks in one or more of the plurality of candidate web pages that point to the web page,

a match between anchor text associated with the outlinks that point to the web page and a name of the business,

a match between a title of the web page and the name of the business,

a number of geographic locations identified in the web page, or

a match between a domain name associated with the web page and the name of the business, and

the one or more instructions to determine the authoritative score for each web page of the plurality of candidate web pages, when executed by the processor, further cause the processor to;

weight and combine a respective plurality of scores associated with the respective plurality of signals for each candidate web page to determine the authoritative score for each candidate web page; and

one or more instructions which, when executed by the processor, cause the processor to identify a particular web page of the plurality of candidate web pages as an authoritative web page for the business,

the particular web page having a highest authoritative score of the authoritative scores for the plurality of candidate web pages.

15. The non-transitory computer-readable medium of claim 14, where the one or more instructions to identify the plurality of candidate web pages further include:

one or more instructions to analyze web pages in a web page corpus to identify snippets of text that include information associated with the business; and

one or more instructions to identify web pages that include the snippets of text as candidate web pages.

16. The non-transitory computer-readable medium of claim 15, where the information associated with the business includes at least one of:

at least a portion of an address associated with a location of the business,

at least a portion of a telephone number associated with the business, or

at least a portion of the name of the business.

17. The non-transitory computer-readable medium of claim 15, where the one or more instructions to identify the plurality of candidate web pages further include:

one or more instructions to identify web pages that are linked to the web pages that include the snippets of text as candidate web pages.

18. The non-transitory computer-readable medium of claim 17, where the one or more instructions to identify the plurality of candidate web pages further include:

one or more instructions to identify one or more geographic addresses, that are not associated with a location of the business, in the plurality of candidate web pages, and

one or more instructions to identify additional web pages in the web page corpus that are also associated with the one or more geographic addresses as candidate web pages.

19. The non-transitory computer-readable medium of claim 14, where the one or more instructions to determine the authoritative score for each web page of the plurality of candidate web pages further include:

one or more instructions to increase the authoritative score for each web page of the plurality of candidate web pages when the number of geographic locations associated with the one of the plurality of candidate web pages is less than a threshold.

20. The non-transitory computer-readable medium of claim 14, where the one or more instructions to identify the plurality of candidate web pages include:

one or more instructions to remove one or more portions of an address of a particular web page included in the plurality of candidate web pages to identify one or more additional web pages as candidate web pages.