Online advertising valuation apparatus and method

Abstract

A method comprising for determining the value provided to an advertiser through the advertiser's online advertisements. The method may begin with providing an online advertisement linked to content. A macro-context may be obtained that quantifies the affinity of the content for a plurality of domains corresponding to a semantic space. A personalization vector may be obtained that quantifies the affinity of at least one aspect of a user for the plurality of domains. The user may then select the online advertisement. A match value may be calculated that quantifies the similarity between the personalization vector and the macro-context. The advertiser may then be charged a monetary amount for the user's “click.” The monetary amount may be based, at least in part, on the match value.

What is claimed and desired to be secured by United States Letters Patent is: 
     
         21 . A computer-implemented method for certifying that a user who is viewing content has viewed substantially similar content at a previous interval of time comprising:
 classifying content using a macro-context vector, wherein the macro-context vector specifies classification of the content within a domain;   assigning a personalization vector to the user based on browser history;   multiplying the personalization vector and the macro-context vector to receive a match value;   receiving, from a content provider, a threshold; and   comparing the match value to the threshold, wherein if the match value is greater than the threshold, generating a certification that the user who is viewing content has viewed substantially similar content at a previous interval of time.   
     
     
         22 . The computer-implemented method of  claim 21 , wherein the content provider is an advertiser. 
     
     
         23 . The computer-implemented method of  claim 21 , wherein the macro-context vector comprises a universal resource locator. 
     
     
         24 . The computer-implemented method of  claim 21 , wherein the personalization vector comprises an aggregation of a plurality of vectors from viewed web pages. 
     
     
         25 . The computer-implemented claim of  21 , wherein the multiplication of the personalization vector and the macro-context vector comprises a dot product. 
     
     
         26 . The computer implemented method of  claim 21 , further comprising:
 providing the match value to the content provider.   
     
     
         27 . The computer-implemented method of  claim 21 , further comprising:
 based on the generation of the certification, determining a quality of visitors to a web page.   
     
     
         28 . The computer-implemented method of  claim 21 , further comprising:
 storing the match value in a repository.   
     
     
         29 . The computer implemented method of  claim 28 , further comprising:
 retrieving the match value from the repository; and   based on the match value, customizing the content for the user.   
     
     
         30 . The computer-implemented method of  claim 21 , further comprising:
 implementing a content auction, wherein a price is determined for a keyword that does not currently have a bid based on the certification.   
     
     
         31 . A computer-implemented method for representing a domain as a set of slots to be filled comprising:
 generating a list of assertions answered by text within the domain to specify the domain itself;   verifying whether the list of text answers relate to a specific question by extracting content of interest to the domain;   based on the verification, storing a result in a database; and   generating a report based on the stored result.   
     
     
         32 . The computer-implemented method of  claim 31 , wherein the list of assertions specifies differences between domains. 
     
     
         33 . The computer-implemented method of  claim 31 , wherein the list of assertions is matched to a previously generated list of assertions. 
     
     
         34 . The computer-implemented method of  claim 31 , further comprising:
 matching the generated list of assertions with the domain.   
     
     
         35 . The computer-implemented method of  claim 31 , further comprising:
 selecting from the list of assertions for the domain; and   creating relationships between the list of assertions answered by text.   
     
     
         36 . The computer-implemented method of  claim 35 , wherein the relationships comprise at least one of adjacency, font size, paragraph length, verb type coexistence, voice, tense, keywords, key phrases or numerical values. 
     
     
         37 . A system comprising:
 a tangible computer-readable storage device comprising instructions; and one or more processors coupled to the tangible computer-readable storage device and configured to execute the instructions to perform operations comprising:   classifying content using a macro-context vector, wherein the macro-context vector specifies classification of the content within a domain;   assigning a personalization vector to the user based on browser history, wherein the personalization vector comprises an aggregation of macro-context vectors;   multiplying the personalization vector and the macro-context vector to receive a match value;   receiving, from a content provider, a threshold; and   comparing the match value to the threshold, wherein if the match value is greater than the threshold, generating a certification that the user who is viewing content has viewed substantially similar content at a previous interval of time.   
     
     
         38 . The system of  claim 37 , the operations further comprising:
 storing the match value in a repository.   
     
     
         39 . The system of  claim 38 , the operations further comprising:
 retrieving the match value from the repository; and   based on the match value, customizing the content for the user.   
     
     
         40 . The system of  claim 37 , the operations further comprising:
 implementing a content auction, wherein a price is determined for a keyword that does not currently have a bid based on the certification.