Associating features with entities, such as categories of web page documents, and/or weighting such features

• US 9,852,225 B2
• Filed: 04/23/2015
• Issued: 12/26/2017
• Est. Priority Date: 12/30/2004
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Koningstein, Ross, Lawrence, Stephen, Spitkovsky, Valentin
• Primary Examiner(s): ALAM, SHAHID AL
• Application Number:
  US14/694,534
• Publication Number:
  US 20150317679A1
• Time in Patent Office: 978 Days
• Field of Search: 707723, 705 1441
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9536 Search customisation based on social or collaborative filtering
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0242 Determining effectiveness of advertisements
• CPC Class Codes: G06Q 30/0256 User search
• CPC Class Codes: G06Q 30/0277 Online advertisement

Abstract

Features that may be used to represent relevance information (e.g., properties, characteristics, etc.) of an entity, such as a document or concept for example, may be associated with the document by accepting an identifier that identifies a document; obtaining search query information (and/or other serving parameter information) related to the document using the document identifier, determining features using the obtained query information (and/or other serving parameter information), and associating the features determined with the document. Weights of such features may be similarly determined. The weights may be determined using scores. The scores may be a function of one or more of whether the document was selected, a user dwell time on a selected document, whether or not a conversion occurred with respect to the document, etc. The document may be a Web page. The features may be n-grams. The relevance information of the document may be used to target the serving of advertisements with the document.

Claims

1. A method comprising:

accessing search queries that were previously submitted by users and a corresponding set of search results that were previously presented in search results pages for each of the search queries, each search result in the set of the search results including a link to a corresponding web page;

identifying, for a given web page that was previously accessed through user interaction with links included in search results from the set of search results, a set of the search queries that were previously used to provide search results pages that included the links with which the user interactions occurred to access the given web page;

determining, for each given search query in the set of search queries, a feature score based on a user dwell time at the given web page following user interaction with the links to the given web page that were provided in the search results pages for the given search query, including assigning a higher scores for higher dwell times at the given web page;

storing, based on the determined features scores, one or more of the search queries from the identified set of search queries as keywords for the given web page, including storing the given query having a best feature score as a given keyword for the given web page;

receiving a request for content to be displayed with the given web page;

selecting content for display with the given web page based on the given keyword for the given web page matching terms used to distribute the content; and

transmitting, via a network, the selected content for display with the given web page responsive to the request for content.

2. The method of claim 1, wherein at least one search query in the set of search queries comprises one or more n-grams.

3. The method of claim 2, comprising determining a weight for each n-gram based on a number of times links to the given web page are selected when the n-gram is included in the at least one search query that was used to present a search results page that includes the links to the given web page.

4. The method of claim 3, wherein determining the weight comprises determining the weight for each n-gram based, at least in part, on a number of conversions that occur after selections of the link for the given web page.

5. The method of claim 3, comprising filtering one or more n-grams from the search queries based on the weight associated with the n-gram being below a predetermined value.

6. A system comprising:

accessing search queries that were previously submitted by users and a corresponding set of search results that were previously presented in search results pages for each of the search queries, each search result in the set of the search results including a link to a corresponding web page;

identifying, for a given web page that was previously accessed through user interaction with links included in search results from the set of search results, a set of the search queries that were previously used to provide search results pages that included the links with which the user interactions occurred to access the given web page;

determining, for each given search query in the set of search queries, a feature score based on a user dwell time at the given web page following user interaction with the links to the given web page that were provided in the search results pages for the given search query, including assigning a higher scores for higher dwell times at the given web page;

storing, based on the determined features scores, one or more of the search queries from the identified set of search queries as keywords for the given web page, including storing the given query having a best feature score as a given keyword for the given web page;

receiving a request for content to be displayed with the given web page;

selecting content for display with the given web page based on the given keyword for the given web page matching terms used to distribute the content; and

transmitting, via a network, the selected content for display with the given web page responsive to the request for content.

7. The system of claim 6, wherein at least one search query in the set of search queries comprises one or more n-grams.

8. The system of claim 7, comprising determining a weight for each n-gram based on a number of times links to the given web page are selected when the n-gram is included in the at least one search query that was used to present a search results page that includes the links to the given web page.

9. The system of claim 8, wherein determining the weight comprises determining the weight for each n-gram based, at least in part, on a number of conversions that occur after selections of the link for the given web page.

10. The system of claim 8, comprising filtering one or more n-grams from the search queries based on the weight associated with the n-gram being below a predetermined value.

11. A computer-readable storage device storing instructions that, when executed by one or more data processors, cause the one or more data processors to perform operations comprising:

accessing search queries that were previously submitted by users and a corresponding set of search results that were previously presented in search results pages for each of the search queries, each search result in the set of the search results including a link to a corresponding web page;

identifying, for a given web page that was previously accessed through user interaction with links included in search results from the set of search results, a set of the search queries that were previously used to provide search results pages that included the links with which the user interactions occurred to access the given web page;

determining, for each given search query in the set of search queries, a feature score based on a user dwell time at the given web page following user interaction with the links to the given web page that were provided in the search results pages for the given search query, including assigning a higher scores for higher dwell times at the given web page;

storing, based on the determined features scores, one or more of the search queries from the identified set of search queries as keywords for the given web page, including storing the given query having a best feature score as a given keyword for the given web page;

receiving a request for content to be displayed with the given web page;

selecting content for display with the given web page based on the given keyword for the given web page matching terms used to distribute the content; and

transmitting, via a network, the selected content for display with the given web page responsive to the request for content.

12. The computer-readable storage device of claim 11, wherein at least one search query in the set of search queries comprises one or more n-grams.

13. The computer-readable storage device of claim 12, comprising determining a weight for each n-gram based on a number of times links to the given web page are selected when the n-gram is included in the at least one search query that was used to present a search results page that includes the links to the given web page.

14. The computer-readable storage device of claim 13, wherein determining the weight comprises determining the weight for each n-gram based, at least in part, on a number of conversions that occur after selections of the link for the given web page.

15. The computer-readable storage device of claim 13, comprising filtering one or more n-grams from the search queries based on the weight associated with the n-gram being below a predetermined value.