Method and system for ranking search content

Abstract

The present teaching relates to ranking search content. In one example, a plurality of documents is received to be ranked with respect to a query. Features are extracted from the query and the plurality of documents. The plurality of documents is ranked based on a ranking model and the extracted features. The ranking model is derived to remove one or more documents from the plurality of documents that are less relevant to the query and order remaining documents based on their relevance to the query. The ordered remaining documents are provided as a search result with respect to the query.

We claim: 
     
         1 . A method for training a model for ranking search content, comprising:
 obtaining, by a computing device, training data comprising a plurality of training data samples each of which includes a pair of a query and a corresponding uniform resource locator (URL);   obtaining, for each of the plurality of training data samples, features characterizing:
 relevance between the query and content associated with the corresponding URL, and 
 user activities related to the corresponding URL; 
   accessing assessment data associated with each of the training data samples, indicating a level of relevance between the corresponding pair of query and URL; and   training the model based on the training data samples as well as the features and the assessment data associated therewith.   
     
     
         2 . The method of  claim 1 , wherein the plurality of query/URL pairs are classified into:
 a first group including query/URL pairs each of which has a first score representing a higher relevance between query and URL in the pair, and   a second group including query/URL pairs each of which has a second score representing a lower relevance between query and URL in the pair.   
     
     
         3 . The method of  claim 2 , wherein the training is based on the first and second scores. 
     
     
         4 . The method of  claim 1 , wherein the user activities comprise one or more of: user click behavior regarding a URL and browsing behavior regarding a URL. 
     
     
         5 . The method of  claim 1 , wherein the relevance between the query and content associated with the corresponding URL comprises textual relevance between the query and content associated with the corresponding URL. 
     
     
         6 . The method of  claim 1 , wherein the training data further comprises user information of a user who submitted the query. 
     
     
         7 . The method of  claim 6 , wherein the user information comprises the user's demographic information. 
     
     
         8 . A non-transitory, computer-readable medium having information recorded thereon for training a model for ranking search content, wherein the information, when read by a machine, causes the machine to perform operations comprising:
 obtaining training data comprising a plurality of training data samples each of which includes a pair of a query and a corresponding uniform resource locator (URL);   obtaining, for each of the plurality of training data samples, features characterizing:
 relevance between the query and content associated with the corresponding URL, and 
 user activities related to the corresponding URL; 
   accessing assessment data associated with each of the training data samples, indicating a level of relevance between the corresponding pair of query and URL; and   training the model based on the training data samples as well as the features and the assessment data associated therewith.   
     
     
         9 . The medium of  claim 8 , wherein the plurality of query/URL pairs are classified into:
 a first group including query/URL pairs each of which has a first score representing a higher relevance between query and URL in the pair, and   a second group including query/URL pairs each of which has a second score representing a lower relevance between query and URL in the pair.   
     
     
         10 . The medium of  claim 9 , wherein the training is based on the first and second scores. 
     
     
         11 . The medium of  claim 8 , wherein the user activities comprise one or more of: user click behavior regarding a URL and browsing behavior regarding a URL. 
     
     
         12 . The medium of  claim 8 , wherein the relevance between the query and content associated with the corresponding URL comprises textual relevance between the query and content associated with the corresponding URL. 
     
     
         13 . The medium of  claim 8 , wherein the training data further comprises user information of a user who submitted the query. 
     
     
         14 . The medium of  claim 13 , wherein the user information comprises the user's demographic information. 
     
     
         15 . A system for training a model for ranking search content, the system comprising:
 memory storing computer program instructions; and   one or more processors that, in response to executing the computer program instructions, effectuate operations comprising:   obtaining training data comprising a plurality of training data samples each of which includes a pair of a query and a corresponding uniform resource locator (URL);   obtaining, for each of the plurality of training data samples, features characterizing:
 relevance between the query and content associated with the corresponding URL, and 
 user activities related to the corresponding URL; 
   accessing assessment data associated with each of the training data samples, indicating a level of relevance between the corresponding pair of query and URL; and   training the model based on the training data samples as well as the features and the assessment data associated therewith.   
     
     
         16 . The system of  claim 15 , wherein the plurality of query/URL pairs are classified into:
 a first group including query/URL pairs each of which has a first score representing a higher relevance between query and URL in the pair, and   a second group including query/URL pairs each of which has a second score representing a lower relevance between query and URL in the pair.   
     
     
         17 . The system of  claim 16 , wherein the training is based on the first and second scores. 
     
     
         18 . The system of  claim 15 , wherein the user activities comprise one or more of: user click behavior regarding a URL and browsing behavior regarding a URL. 
     
     
         19 . The system of  claim 15 , wherein the relevance between the query and content associated with the corresponding URL comprises textual relevance between the query and content associated with the corresponding URL. 
     
     
         20 . The system of  claim 15 , wherein the training data further comprises user information of a user who submitted the query.