Structured informational link annotations

Abstract

Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium. A method includes: for each of a plurality of content items in an inventory of content items: identifying an entity associated with the content item and a plurality of page types for a vertical associated with a product or service described in the content item; locating a plurality of informational pages associated with the entity; classifying each informational page based on the page types; identifying queries associated with the entity, wherein a query is used as a selection criteria for delivering one or more content items associated with the entity; for each informational page of the plurality of informational pages determining relevant queries from the identified queries; and storing in a data structure an association between the content item, data associated with the relevant queries and associated informational pages.

What is claimed is: 
     
         1 . A method comprising
 for each of a plurality of content items in an inventory of content items:
 identifying an entity associated with the content item and a plurality of page types for a vertical associated with a product or service described in the content item; 
 locating a plurality of informational pages associated with the entity; 
 classifying each informational page based on the page types; 
 identifying queries associated with the entity, wherein a query is used as a selection criteria for delivering one or more content items associated with the entity; 
 for each informational page of the plurality of informational pages determining relevant queries from the identified queries; and 
 storing in a data structure an association between the content item, data associated with the relevant queries and associated informational pages. 
   
     
     
         2 . The method of  claim 1  wherein the entity is a content sponsor. 
     
     
         3 . The method of  claim 1  wherein determining relevant queries from the identified queries includes determining one or more keywords for a given informational page and identifying a relevant query includes finding one or more queries that include the determined one or more keywords. 
     
     
         4 . The method of  claim 1  further comprising
 receiving a request for content; 
 identifying a content item responsive to the request; 
 identifying one or more informational pages associated with the content item based at least in part on a content sponsor associated with the content item and terms included in the request and/or the landing page associated with the content item; 
 generating a link for an informational page of the one or more informational pages; 
 augmenting the content item with the generated link; and 
 providing the augmented content item responsive to the request. 
 
     
     
         5 . The method of  claim 1  wherein identifying a plurality of page types for a vertical associated with a product or service described in the content item includes retrieving a set of page types from a database based on the vertical. 
     
     
         6 . The method of  claim 1  wherein identifying the plurality of page types includes evaluating a corpus of documents associated with a document sponsor to determine the page types. 
     
     
         7 . The method of  claim 6  wherein evaluating the corpus includes evaluating titles of documents in the corpus including extracting n-grams from the titles that do not include a product, service or brand name associated with the content sponsor and using the extracted n-grams to identify the page types. 
     
     
         8 . The method of  claim 1  further comprising determining which of the identified page types to assign to a content item by evaluating one or more of URL (Uniform Resource Locator) patterns or title n-grams. 
     
     
         9 . The method of  claim 2  wherein locating a plurality of informational pages associated with the content sponsor includes receiving a list of informational pages associated with the content sponsor. 
     
     
         10 . The method of  claim 2  wherein locating a plurality of informational pages associated with the content sponsor includes evaluating a corpus of documents associated with the content sponsor to identify the plurality of informational pages. 
     
     
         11 . The method of  claim 2  wherein locating a plurality of informational pages associated with the content sponsor includes identifying a set of URLs that constitute informational links for the page types of the plurality of page types chosen from a total set of URLs associated with the content sponsor. 
     
     
         12 . The method of  claim 1  wherein the page types are associated with informational needs of a user selected from how-to, buying guide, reviews, product walkthrough, product gallery, customer reviews, question and answer, live chat, technical specifications, technical support, top lists, or side-by-side comparisons. 
     
     
         13 . The method of  claim 4  wherein identifying one or more informational pages includes selecting, based on one or more criteria, one or more informational pages from the identified informational pages and generating an informational link based on a selected informational page. 
     
     
         14 . The method of  claim 13  further comprising determining a task associated with the request, wherein the task is located along a path toward conversion and wherein the criteria relates to furthering the user along the path toward conversion. 
     
     
         15 . The method of  claim 13  further comprising assembling the generated informational links in an order and providing the assembled informational links along with the content item responsive to the request. 
     
     
         16 . The method of  claim 4  further comprising determining a subset of the informational pages to present based on one or more criteria and augmenting includes presenting informational links associated with the subset. 
     
     
         17 . The method of  claim 16  wherein the criteria are based on a function of a task that is inferred that the user is performing related to the request. 
     
     
         18 . The method of  claim 4  further comprising automatically constructing a title for the generated link. 
     
     
         19 . The method of  claim 18  wherein the title is automatically constructed based on one or more criteria including page type and screen space available. 
     
     
         20 . The method of  claim 2  wherein locating a plurality of informational pages associated with the content sponsor includes submitting a request to a search system for the search system to locate informational pages that include one or more page type n-grams and receiving information from the search system identifying the located informational pages. 
     
     
         21 . A method comprising:
 identifying a content item and an entity associated with the content item;   identifying a plurality of informational pages associated with the entity;   determining a plurality of page types associated with informational content that are related to the entity;   classifying each informational page based on the page types;   determining which of the informational pages relate to the content item based on one or more criteria; and   storing in a database an association between the content item, the content sponsor, and related informational pages based on page type.   
     
     
         22 . The method of  claim 21  wherein the one or more criteria include selection criteria which are used to determine when to deliver the content item, content of the content item, or content of a landing page associated with the content item. 
     
     
         23 . The method of  claim 21  wherein the entity is a content sponsor. 
     
     
         24 . A system comprising:
 a link and criteria identification system;   an annotation serving system;   an annotation rendering system; and   a content selector;   wherein the content selector is configured to:
 receive a request for content; and 
 identify a content item responsive to the request; and 
   wherein the link and criteria identification system is configured to identify one or more informational pages associated with the content item based at least in part on a content sponsor associated with the content item and terms included in the request and/or the landing page associated with the content item; and   wherein the annotation serving system is configured to:
 generate a link for an informational page of the one or more informational pages; and 
 augment the content item with the generated link; and 
   wherein the annotation rendering system is configured to provide the augmented content item responsive to the request.   
     
     
         25 . A computer program product tangibly embodied in a computer-readable storage device and comprising instructions that, when executed by a processor, cause the processor to:
 for each of a plurality of content items in an inventory of content items:
 identify an entity associated with the content item and a plurality of page types for a vertical associated with a product or service described in the content item; 
 locate a plurality of informational pages associated with the entity; 
 classify each informational page based on the page types; 
 identify queries associated with the entity, wherein a query is used as a selection criteria for delivering one or more content items associated with the entity; 
 for each informational page of the plurality of informational pages determine relevant queries from the identified queries; and 
 store in a data structure an association between the content item, data associated with the relevant queries and associated informational pages.