Method and system for sponsored search results placement in a search results page

Abstract

The present teaching relates to placing sponsored search results based on correlation of the sponsored search results. A search query is first received at a search engine from a user. One or more keywords are further extracted horn the search query. A plurality of sponsored search results related to the one or more keywords are received in response to the search query. The placement of the plurality of sponsored search results are further determined based on correlation of the plurality of sponsored search results, and a search results page containing the plurality of sponsored search results are presented to the user.

We claim: 
     
         1 . A method, implemented on at least one computing device each of which has at least one processor, storage, and a communication platform connected to a network for placing sponsored search results in a search results page, the method comprising:
 receiving a search query from a user;   extracting one or more keywords from the search query,   receiving a plurality of sponsored search results related to the one or more keywords;   determining placement of the plurality of sponsored search results based on correlations of the plurality of sponsored search results; and   presenting to the user, a search results page containing the plurality of sponsored search results.   
     
     
         2 . The method of  claim 1 , further comprising:
 obtaining profile information of the user;   retrieving user behaviors from a database based on the profile information; and   determining the placement of the plurality of sponsored search results based on the user behaviors.   
     
     
         3 . The method of  claim 2 , further comprising:
 estimating a user likelihood factor for each sponsored search result based on the user behaviors; and   determining the placement of the plurality of sponsored search results based on the user likelihood factors, wherein   the user likelihood factor indicates a chance that the user clicks the sponsored search result.   
     
     
         4 . The method of  claim 1 , further comprising:
 determining the placement of the plurality of sponsored search results based on qualities of the plurality of sponsored search results, which includes:
 obtaining a plurality of quality factors associated with each sponsored search result; 
 estimating a quality grade for each sponsored search result based on the plurality of quality factors; and 
 determining the placement of the plurality of sponsored search results based on the quality grades. 
   
     
     
         5 . The method of  claim 1 , further comprising:
 obtaining a creative associated with each sponsored search result;   estimating a correlation factor for each sponsored search result based on the creatives; and   determining the placement of the plurality of sponsored search results based on the correlation factors, wherein   the correlation factor indicates an interaction level between a sponsored search result and other sponsored search results to be placed together on the search results page.   
     
     
         6 . The method of  claim 1 , further comprising:
 predicting a number of clicks on each sponsored search result based on user behaviors, qualities of the plurality of sponsored search results, and the correlations of the plurality of sponsored search results using a machine learning model; and   determining the placement of the plurality of sponsored search results based on the predicted numbers of clicks.   
     
     
         7 . A method, implemented on at least one computing device each of which has at least one processor, storage, and a communication platform connected to a network for placing sponsored search results in a search results page, the method comprising:
 receiving a search query from a user;   extracting one or more keywords from the search query;   receiving a plurality of sponsored search results related to the one or more keywords;   determining placement of the plurality of sponsored search results based on user behaviors and qualities of the plurality of sponsored search results; and   presenting to the user, a search results page containing the plurality of sponsored search results.   
     
     
         8 . The method of  claim 7 , further comprising:
 predicting a number of clicks on each sponsored search result based on user behaviors and qualities of the plurality of sponsored search results using a machine learning model; and   determining the placement of the plurality of sponsored search results based on the predicted numbers of clicks.   
     
     
         9 . A system having at least one processor, storage, and a communication platform connected to a network for placing sponsored search results in a search results page, the system comprising:
 a user interfacing module configured to receive a search query from a user;   a keywords extractor configured to extract one or more keywords from the search query;   a receiving module configured to receive a plurality of sponsored search results related to the one or more keywords;   a decision module configured to determine placement of the plurality of sponsored search results based on correlations of the plurality of sponsored search results; and   a presenting module configured to present to the user, a search results page containing the plurality of sponsored search results.   
     
     
         10 . The system of  claim 8 , further comprising:
 a user profile collector configured to obtain profile information of the user; and   a user behavior analyzer configured to retrieve user behaviors from a database based on the profile information, wherein   the decision module is further configured to determine the placement of the plurality of sponsored search results based on the user behaviors.   
     
     
         11 . The system of  claim 10 , wherein the user behavior analyzer is further configured to estimate a user likelihood factor for each sponsored search result based on the user behaviors, the user likelihood factor indicating a chance that the user clicks the sponsored search result. 
     
     
         12 . The system of  claim 8 , wherein the decision module is further configured to determine the placement of the plurality of sponsored search results based on qualities of the plurality of sponsored search results, which includes:
 obtaining a plurality of quality factors associated with each sponsored search result;   estimating a quality grade for each sponsored search result based on the plurality of quality factors; and   determining the placement of the plurality of sponsored search results based on the quality grades.   
     
     
         13 . The system of  claim 8 , further comprising
 a creative extractor configured to obtain a creative associated with each sponsored search result; and   a correlation analyzer configured to estimate a correlation factor for each sponsored search result based on the creatives, the correlation factor indicating an interaction level between a sponsored search result and other sponsored search results to be placed together on the search results page, wherein   the decision module is further configured to determine the placement of the plurality of sponsored search results based on the correlation factors.   
     
     
         14 . The system of  claim 8 , further comprising;
 a clicks prediction module configured to predict a number of clicks on each sponsored search result based on user behaviors, qualities of the plurality of sponsored search results, and the correlations of the plurality of sponsored search results using a machine learning model, wherein   the decision module is further configured to determine the placement of the plurality of sponsored search results based on the predicted numbers of clicks.   
     
     
         15 . A non-transitory machine-readable medium having information recorded thereon for placing sponsored search results in a search results page, wherein the information, when read by the machine, causes the machine to perform the following;
 receiving a search query from a user;   extracting one or more keywords from the search query;   receiving a plurality of sponsored search results related to the one or more keywords;   determining placement of the plurality of sponsored search results based on correlations of the plurality of sponsored search results; and   presenting to the user, a search results page containing the plurality of sponsored search results.   
     
     
         16 . The non-transitory machine-readable medium of  claim 15 , wherein the information, when read by the machine, causes the machine to further perform the following:
 obtaining profile information of the user;   retrieving user behaviors from a database based on the profile information; and   determining the placement of the plurality of sponsored search results based on the user behaviors.   
     
     
         17 . The non transitory machine-readable medium of  claim 16 , wherein the information, when read by the machine, causes the machine to further perform the following:
 estimating a user likelihood factor for each sponsored search result based on the user behaviors;   determining the placement of the plurality of sponsored search results, based on the user likelihood factors, wherein   the user likelihood factor indicates a chance that the user clicks the sponsored search result.   
     
     
         18 . The non-transitory machine-readable medium of  claim 15 , wherein the information, when read by the machine, causes the machine to further perform the following:
 determining the placement of the plurality of sponsored search results based on qualities of the plurality of sponsored search results, which includes obtaining a plurality of quality factors associated with each sponsored search result;   estimating a quality uncle for each sponsored search result based on the plurality of quality factors; and   determining the placement of the plurality of sponsored search results based on the quality grades.   
     
     
         19 . The non-transitory machine-readable medium of  claim 15 , wherein the information, when read by the machine, causes the machine to further perform the following:
 obtaining a creative associated with each sponsored search result;   estimating a correlation factor for each sponsored search result based on the creatives; and   determining the placement of the plurality of sponsored search results based on the correlation factors, wherein   the correlation factor indicates an interaction level between a sponsored search result and other sponsored search results to be placed together on the search results page.   
     
     
         20 . The non-transitory machine-readable medium of  claim 15 , wherein the information, when read by the machine, causes the machine to further perform the following:
 predicting a number of clicks on each sponsored search result based on user behaviors, qualities of the plurality of sponsored search results, and the correlations of the plurality of sponsored search results using a machine learning model; and   determining the placement of the plurality of sponsored search results based on the predicted number of clicks.