Identifying missing search phrases

Abstract

In various embodiments, approaches for identifying missing search phrases are described. Search terms are behaviorally associated with an item in the electronic repository. If a search of the electronic repository fails to surface the item, the at least one search term can be identified as a missing search phrase for the item.

1 . A method comprising:
 determining, by at least one computing device, that a degree of behavioral association between a search term and an item meets a threshold;   determining, by the at least one computing device, that a search of a repository for the search term surfaces the item with a corresponding relevance level that fails to meet a relevance threshold; and   in response to determining that the degree of behavioral association between the search term and the item meets the threshold and determining that the search surfaces the item with the corresponding relevance level that fails to meet the relevance threshold, performing, by the at least one computing device, at least one remedial measure.   
     
     
         2 . The method of  claim 1 , wherein determining that the degree of behavioral association between the search term and the item comprises calculating a weighted value based at least in part upon behavior data of a user. 
     
     
         3 . The method of  claim 1 , wherein the degree of behavioral association between the search term and the item is based upon purchasing activity of a population of users submitting the search term to a search engine and purchasing at least one item from an electronic commerce system after submitting the search term. 
     
     
         4 . The method of  claim 3 , wherein submitting the search term to the search engine and purchasing the at least one item from the electronic commerce system occurs during a respective user session of each of the users among the population of users. 
     
     
         5 . The method of  claim 1 , wherein the degree of behavioral association between the search term and the item is based upon browsing data of a population of users submitting the search term to a search engine and browsing to the item after submitting the search term during a respective user session. 
     
     
         6 . The method of  claim 1 , wherein the at least one remedial measure comprises at least one of associating the search term with the item, modifying metadata associated with the item to include the search term, or modifying a search algorithm used to perform the search. 
     
     
         7 . The method of  claim 1 , further comprising modifying the threshold to tune the degree of the behavioral association. 
     
     
         8 . The method of  claim 1 , wherein the repository is an electronic repository of an electronic commerce system. 
     
     
         9 . The method of  claim 1 , wherein the degree of the behavioral association between the search term and the item is based upon whether a threshold number of users submit the search term and follow a hyperlink corresponding to the item during a user session. 
     
     
         10 . The method of  claim 1 , wherein taking the at least one remedial measure further comprises adding, by the at least one computing device, the search term to a list of search terms associated with the item in the repository. 
     
     
         11 . The method of  claim 1 , wherein determining that the search of the repository for the search term surfaces the item with a corresponding relevance level that fails to meet the relevance threshold further comprises determining that the search term fails to appear in at least one attribute associated with the item in the repository. 
     
     
         12 . The method of  claim 1 , wherein determining that the search of the repository for the search term surfaces the item with a corresponding relevance level that fails to meet the relevance threshold further comprises performing a natural language search of the repository with the search term and determining that the natural language search fails to associate the search term with the item. 
     
     
         13 . A system, comprising:
 at least one computing device; and   a search application executed on the at least one computing device, the search application, when executed, causing the at least one computing device to at least:
 determine that a search of an electronic repository for a search term surfaces an item that fails to meet a relevance threshold with respect to the search term; 
 determine that a behavioral association exists between the item and the search term based at least in part upon behavior of at least one user with respect to interactions of the at least one user with the at least one computing device; and 
 in response to the determination that the behavioral association exists, take at least one remedial measure. 
   
     
     
         14 . The system of  claim 13 , wherein the at least one remedial measure comprises at least one of associating the search term with the item, modifying metadata associated with the item, or modifying a search algorithm. 
     
     
         15 . The system of  claim 13 , wherein the search application further causes the at least one computing device to at least associate, in response to the determination that the behavioral association exists, the search term with at least one other item in the electronic repository that is related to the item. 
     
     
         16 . The system of  claim 15 , wherein the interactions comprise a user searching the search term in a user session followed by at least one of adding the item to a shopping cart or viewing the item. 
     
     
         17 . A non-transitory computer readable medium embodying a computer program that, when executed by at least one computing device, causes the at least computing device to at least:
 determine that a degree of behavioral association between a search term and a particular item meets a threshold, the degree of behavioral association based at least in part upon behavior data of a plurality of users with respect to interactions of the plurality of users with the particular item;   determine that the particular item is in a plurality of search results at a position that fails to meet a relevance threshold, wherein the plurality of search results is based at least in part on the search term; and   in response to determining that the degree of behavioral association between the search term and the particular item meets the threshold and determining that the particular item is in the plurality of search results at the position that fails to meet the relevance threshold, perform at least one remedial measure to improve the position of the particular item within the search results.   
     
     
         18 . The non-transitory computer readable medium of  claim 17 , wherein the at least one remedial measure comprises associating the search term with the particular item, modifying product data associated with the particular item, writing the search term to a log as a potential missing search term, or modifying a search algorithm used to generate the plurality of search results. 
     
     
         19 . The non-transitory computer readable medium of  claim 17 , wherein a determination that the degree of behavioral association between the search term and the particular item meets the threshold is performed by calculating a weighted value based at least in part upon behavior data of a user. 
     
     
         20 . The non-transitory computer readable medium of  claim 17 , wherein the at least one remedial measure comprises adding the search term to metadata associated with the particular item.