Method and system for identifying information relevant to content

• US 8,782,074 B1
• Filed: 03/22/2010
• Issued: 07/15/2014
• Est. Priority Date: 06/20/2003
• Status: Expired due to Term

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Agrawal, Ashish, Frederick, Robert
• Primary Examiner(s): Alam, Hosain
• Assistant Examiner(s): Ahluwalia, Navneet K
• Application Number:
  US12/729,115
• Time in Patent Office: 1,576 Days
• Field of Search: 707 1- 10, 707/768, 707/769
• US Class Current: 707/768
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0603 Catalogue ordering
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]

Abstract

A method and system for identifying information to be associated with content of a display page. The system provides a web service through which associates of a vendor can request and receive information such as product data to be displayed on the associate'"'"'s display pages. The system may receive from an associate'"'"'s computer a request for product data that may include content derived from a web page on which the product is to be displayed. Upon receiving the request, the system identifies a query based on popularity of the query among users, executes the query, and provides an associate with information relating to the results of the executed query.

Claims

1. A computer-based method for associating an item with submitted web page content and providing information about the item, the method comprising:

under the control of one or more computer systems configured with executable instructions,

storing, at an information server of an electronic marketplace, a plurality of queries previously submitted by users in relation to a plurality of items including the item, wherein individual previously-submitted queries are stored with a popularity of submission for the individual previously-submitted queries;

responsive to a link-selection request by a user for a web page of a media web site provided by a media content server, receiving, at the information server of the electronic marketplace, an automatically generated request from the media content server to provide item information based at least in part on the web page content to be displayed on the media web site, the media content server being different than the information server of the electronic marketplace, the web page content submitted in the generated request having been derived by the media content server;

matching, at the information server, the item to the submitted web page content by at least;

identifying a previously-submitted query from the plurality of previously-submitted queries as a matching query based at least in part on a relevance of the matching query to the submitted web page content and the popularity of submission of the matching query;

executing the identified previously-submitted query independent of user input;

identifying a set of one or more items that correspond to the executed query; and

selecting the item from the identified set of one or more items as a matching item; and

providing, asynchronously relative to when the web page content is made available by the media content server, the item information about the matching item to the media content server to be sent to a user device as supplemental content for display on the media web site along with the web page content.

2. The method of claim 1, wherein the plurality of previously-submitted queries correspond to queries submitted by the users to a search engine system that is separated from the media content server by a public communications network.

3. The method of claim 1, wherein the relevance of the matching query to the submitted web page content is based at least in part on one or more phrases in the submitted web page content corresponding to the matching query.

4. The method of claim 1, wherein said identifying the previously-submitted query includes selecting a previously-submitted query that is most popularly submitted.

5. The method of claim 1, wherein the selecting of the item includes:

ranking identified items based at least in part on experiences of users who accessed results of similar queries; and

selecting a high-ranking item as the item that satisfies the identified query.

6. A computer system of an electronic marketplace for providing an item relating to web page content, comprising:

storage for a plurality of previously-submitted queries that have been submitted by users of the electronic marketplace, and indications of popularity of the queries among said users;

at least one processor operable to cause the computer system to at least;

responsive to a link-selection request by a user for a web page of a media web site provided by a media content server, receive an automatically generated request from the media content server to provide information about the item based at least in part on the web page content displayed on the media web site, the media content server being different than the computer system of the electronic marketplace, the web page content submitted in the generated request being derived by the media content server;

match at least a portion of the web page content to the item by at least;

identifying, from the storage, the previously-submitted queries that correspond to at least a portion of the web page content;

selecting an identified previously-submitted query as a matching query based at least in part on a relevance of the matching query to the web page content and a popularity of the matching query; and

identifying the item among a set of items that corresponds to the matching query; and

provide, asynchronously relative to when the web page content is made available by the media content server, the information about the identified item for display on the media web site along with the web page content.

7. The computer system of claim 6, wherein said at least one processor is further operable to identify the information about the item by submitting the selected identified previously-submitted query to a query engine.

8. The computer system of claim 6, wherein said at least one processor is further operable to identify the previously-submitted queries by identifying longest phrases of the web page content that correspond to one or more of the previously-submitted queries.

9. The computer system of claim 6, wherein the popularity of a particular previously-submitted query is based at least in part on purchases by users of the item identified by results of the particular previously-submitted query.

10. The computer system of claim 6, wherein the popularity of a particular previously-submitted query is based at least in part on user requests for information about a product identified in results of a query engine having processed the particular previously-submitted query.

11. A non-transitory computer-readable storage medium containing instructions for controlling a computer system of an electronic marketplace to provide data about an item by a method comprising:

storing a plurality of queries previously submitted to the computer system of the electronic marketplace, wherein individual previously-submitted queries are stored with a popularity of submission for the previously-submitted query;

responsive to a link-selection request by a user for a web page of a media web site provided by a media content server, receiving an automatically generated request from the media content server to provide information based at least in part on web page content to be displayed on the media web site, the media content server being different than the computer system of the electronic marketplace, the web page content submitted in the generated request having been derived by the media content server;

matching the content to the item by at least;

identifying a previously-submitted query from the plurality of previously-submitted queries as a matching query, said identifying based at least in part on a relevance of the matching query to the web page content and the popularity of submission of the matching query; and

selecting the item from a set of one or more items that have been determined to correspond to the matching query; and

providing, asynchronously relative to when the web page content is made available by the media content server, the information about the selected item to be displayed on the media web site along with the web page content.

12. The non-transitory computer-readable storage medium of claim 11, wherein the plurality of previously-submitted queries correspond to queries submitted by the users of the computer system.

13. The non-transitory computer-readable storage medium of claim 11, wherein the relevance of the matching query to the web page content is based at least in part on matching phrases in the web page content to the previously-submitted queries.

14. The non-transitory computer-readable storage medium of claim 11, wherein the identifying of previously-submitted queries includes selecting a particular previously-submitted query that is most popular.

15. The non-transitory computer-readable storage medium of claim 11, wherein the selecting of the item includes:

identifying items that correspond to the matching query;

ranking the identified items based at least in part on experiences of users who accessed results of similar queries; and

selecting a high-ranking item as the item that satisfies the identified query.

16. A computer implemented method for identifying previously-submitted queries related to web page content served by a media content server, comprising:

under the control of one or more computer systems configured with executable instructions,

storing a plurality of previously-submitted queries with corresponding popularity values;

responsive to a link-selection request by a user for a web page of a media web site provided by a media content server, receiving an automatically generated request from the media content server for providing information about an item to be displayed on a user device based at least in part on the web page content to be displayed on the media web site, the media content server being different than the one or more computers, the web page content in the generated request having been derived by the media content server;

responsive to a request by a user for the web page of the media web site provided by the media content server, receiving an automatically generated request from the media content server to provide item information based at least in part on the web page content displayed on the media web site, the media content server being remote with respect to the one or more computer systems, the web page content submitted in the generated request having been derived by the media content server;

responsive to receiving the automatically generated request, identifying phrases of the web page content served by the media content server that match at least one of the previously-submitted queries;

selecting, based at least in part on a length of the identified phrases and the popularity values of corresponding queries, at least one of the identified phrases; and

providing said selected at least one of the identified phrases as representative of the web page content served by the content server.

17. The method of claim 16, further comprising removing at least one noise word from the web page content.

18. The method of claim 16, wherein said identifying includes excluding at least one phrase of the web page content based at least in part on the length of said at least one phrase.

19. The method of claim 16, wherein said storing includes organizing the plurality of previously-submitted queries in a popularity-based query table.

20. The method of claim 19, wherein said identifying includes applying a hash function to potential phrases to reference an index of the popularity-based query table.

21. A non-transitory computer-readable storage medium having stored thereon instructions for causing a computer to perform a method for identifying previously-submitted queries related to web page content served by a content server, the method comprising:

storing a plurality of previously-submitted queries with corresponding popularity values;

responsive to a link-selection request by a user for a web page of a media web site provided by the content server, receiving, by an information server of an electronic marketplace, an automatically generated request from the content server to provide information about an item to be displayed on a user device based at least in part on the web page content to be displayed on the media web site, the media content server being different than the information server of the electronic marketplace, the web page content in the generated request having been derived by the media content server;

responsive to receiving the automatically generated request, identifying phrases of the content served by the content server that match at least one of the previously-submitted queries;

selecting, based at least in part on a length of the identified phrases and the popularity values of corresponding queries, at least one of the identified phrases; and

providing said selected at least one of the identified phrases as representative of the web page content served by the content server.

22. The non-transitory computer-readable storage medium of claim 21, wherein the method further comprises removing at least one noise word from the web page content.

23. The non-transitory computer-readable storage medium of claim 21, wherein said identifying includes excluding at least one phrase of the web page content based at least in part on the length of said at least one phrase.

24. The non-transitory computer-readable storage medium of claim 21, wherein said storing includes organizing the plurality of previously-submitted queries in a popularity-based query table.

25. The non-transitory computer-readable storage medium of claim 24, wherein said identifying includes applying a hash function to potential phrases to reference an index of the popularity-based query table.