Ranking documents based on user behavior and/or feature data

• US 9,305,099 B1
• Filed: 01/10/2012
• Issued: 04/05/2016
• Est. Priority Date: 06/17/2004
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Dean, Jeffrey A., Anderson, Corin, Battle, Alexis
• Primary Examiner(s): Lin, Shew-Fen
• Application Number:
  US13/347,472
• Time in Patent Office: 1,547 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: G06F 40/134 Hyperlinking

Abstract

A system generates a model based on feature data relating to different features of a link from a linking document to a linked document and user behavior data relating to navigational actions associated with the link. The system also assigns a rank to a document based on the model.

Claims

1. A method comprising:

generating, by one or more devices, a rank for a particular document,

generating the rank including;

determining particular feature data associated with a link to the particular document,

the particular feature data identifying one or more attributes of the link,

determining a weight indicating a probability of the link being selected,

the weight being determined based on the particular feature data and selection data,

 

the selection data identifying user behavior relating to links to other documents,

 

the links including one or more links that were selected by user devices and one or more other links that were not selected by the user devices,

 

the weight indicating a higher probability of the link being selected when the particular feature data corresponds to feature data associated with the one or more links than when the particular feature data corresponds to feature data associated with the one or more other links,

 

the feature data associated with the one or more links and the feature data associated with the one or more other links being stored, in a memory associated with the one or more devices, as featured data associated with the links to the other documents,

 

the featured data, associated with the links, identifying;

 

context relating to one or more words before or after the links,

 

words in anchor text associated with the links, and

 

a quantity of the words in the anchor text,

 

the weight being determined based on whether the particular feature data corresponds to the stored feature data associated with the one or more links or whether the particular feature data corresponds to the stored feature data associated with the one or more other links,

the rank being generated based on the weight;

identifying, by the one or more devices, documents associated with a search query,

the documents, associated with the search query, including the particular document; and

providing, by the one or more devices, information associated with the particular document based on;

the search query, and

the generated rank.

2. The method of claim 1, further comprising:

sorting information regarding the documents based on ranks associated with the documents; and

providing the sorted information regarding the documents,

where providing the sorted information regarding the documents includes providing the information associated with the particular document.

3. The method of claim 1, further comprising:

generating rules for a model that ranks certain documents based on information relating to links associated with the certain documents,

the links associated with the certain documents including the links to the other documents,

where the rank is generated using the model.

4. The method of claim 3, where generating the rules for the model includes:

obtaining information identifying attributes of the links to the other documents; and

generating the model based on the obtained information.

5. The method of claim 4, where the information identifying the attributes includes at least one of:

information identifying a quantity of words associated with the links to the other documents, or

information identifying words associated with the links to the other documents.

6. The method of claim 4, where the information identifying the attributes includes at least one of:

information identifying font sizes associated with the links to the other documents, or

information identifying font colors associated with the links to the other documents.

7. The method of claim 4, where the information identifying the attributes includes at least one of:

information identifying positions of the links, to the other documents, in one or more documents, or

information identifying positions of the links, to the other documents, within one or more lists.

8. One or more devices comprising:

one or more memories to store instructions; and

one or more processors to execute the instructions to;

store, in a memory associated with the one or more devices, feature data associated with links to a plurality of documents,

the feature data, associated with the links, identifying;

words in anchor text associated with the links,

a quantity of the words in the anchor text, and

context relating to one or more words before or after the links,

the feature data associated with the links including feature data associated with one or more links that were selected and feature data associated with one or more other links that were not selected, generate a rank for a particular document,

when generating the rank, the one or more processors are to;

determine particular feature data associated with a link to the particular document,

 

the particular feature data identifying one or more attributes of the link,

determine a weight associated with the link,

 

the weight indicating a probability of the link being selected,

 

the weight being determined based on the stored feature data associated with the one or more links that were selected, the stored feature data associated with the one or more other links that were not selected, the particular feature data, and selection data,

 

the selection data identifying user behavior relating to the links,

 

the weight indicating a higher probability of the link being selected when the particular feature data corresponds to the stored feature data associated with the one or more links than when the particular feature data corresponds to the stored feature data associated with the one or more other links,

the rank being generated based on the weight,

identify documents associated with a search query,

the documents including the particular document, and

provide information associated with the particular document based on;

the generated rank, and

the search query.

9. The one or more devices of claim 8, where the one or more processors are further to:

generate rules for a model based on information relating to links associated with certain documents,

the links associated with the certain documents including the links to the plurality of documents,

where the rank is generated using the model.

10. The one or more devices of claim 9, where, when generating the rules for the model, the one or more processors are to obtain the information relating to the links to the plurality of documents,

the obtained information including information identifying attributes of the links to the plurality of documents.

11. The one or more devices of claim 10, where the information identifying the attributes includes at least one of:

information identifying of a quantity of words associated with the links to the plurality of documents, or

information identifying words associated with the links to the plurality of documents.

12. The one or more devices of claim 10, where the information identifying the attributes includes at least one of:

information identifying font sizes of text associated with the links to the plurality of documents,

information identifying font colors of the text associated with the links to the plurality of documents, or

information identifying positions of the links, to the plurality of documents, in particular documents.

13. The one or more devices of claim 10, where the information identifying the attributes includes at least one of:

information indicating whether one of the links, to the plurality of documents, refers to a domain of a document that includes the one of the links, or

information indicating whether the one of the links embeds information identifying another link.

14. The one or more devices of claim 9, where the one or more processors are further to:

determine ranks of the documents associated with the search query using the model;

sort information regarding the documents associated with the search query based on the determined ranks; and

provide the sorted information regarding the documents.

15. A non-transitory computer-readable medium storing instructions, the instructions comprising:

a plurality of instructions which, when executed by one or more devices, cause the one or more devices to;

generate a rank for a particular document,

one or more instructions, of the plurality of instructions, to generate the rank including;

one or more instructions to determine particular feature data associated with a link to the particular document,

 

the particular feature data identifying one or more attributes of the link, and

one or more instructions to determine a weight indicating a probability of the link being selected,

 

the weight being determined based on the particular feature data and selection data,

 

the selection data identifying user behavior relating to links to other documents,

 

the links including one or more links that were selected by user devices and one or more other links that were not selected by the user devices,

 

the weight indicating a higher probability of the link being selected when the particular feature data corresponds to feature data associated with the one or more links than when the particular feature data corresponds to feature data associated with the one or more other links,

 

the feature data associated with the one or more links and the feature data associated with the one or more other links being stored, in a memory associated with the one or more devices, as featured data associated with the links to the other documents,

 

the featured data, associated with the links, identifying;

 

context relating to one or more words before or after the links,

 

words in anchor text associated with the links, and

 

a quantity of the words in the anchor text,

the rank being generated based on the weight;

identify documents associated with a search query,

the documents including the particular document; and

provide information associated with the particular document based on;

the search query, and

the generated rank.

16. The non-transitory computer-readable medium of claim 15, the instructions further comprising:

one or more instructions to generate rules for a model based on information identifying attributes of the links to the other documents,

each of the rules indicating a probability that a link, associated with a respective attribute of the attributes, will be selected; and

one or more instructions to generate the rank for the particular document using the model.

17. The non-transitory computer-readable medium of claim 16, the instructions further comprising one or more instructions to:

determine ranks of the documents associated with the search query using the model;

sort information regarding the documents associated with the search query based on the determined ranks; and

provide the sorted information regarding the documents.

18. The non-transitory computer-readable medium of claim 16, where the information identifying the attributes includes at least one of:

information identifying font sizes of text associated with the links to the other documents, or

information identifying font colors of the text associated with the links to the other documents.

19. The non-transitory computer-readable medium of claim 16, where the information identifying the attributes includes at least one of:

information identifying types of the links to the other documents, or

information identifying topical clusters associated with the links to the other documents.

20. The non-transitory computer-readable medium of claim 16, where the information identifying the attributes includes at least one of:

information identifying positions of the links, to the other documents, in one or more documents, or

information identifying words associated with the links to the other documents.