Application query conversion

Abstract

A set of potential search-query terms can be identified based on empirical queries for apps. For each potential search-query term, a subset of documents within a set of documents can be identified based on apps that users were likely to click on or download following entry of a search query with a comparable or same term. One or more other indicator terms can be identified as being related to the potential search-query term based on the one or more second indicator terms being prevalent within the subset of documents. Upon receipt of a subsequent search query, a search can then be performed using both a term within the search query and one or more related other indicator terms.

What is claimed is: 
     
         1 . A method of defining a data structure to use for expanding a search query requesting relevant software applications from a database storing software applications and a set of documents, each document being associated with a software application, the method comprising:
 identifying, with a server, a set of N potential query terms by analyzing words in the set of documents, N being an integer greater than 100;   defining a co-occurrence matrix defined by an N×N array of matrix elements, wherein each row corresponds to a respective one of the N potential query terms and each column corresponds to a respective one of the N potential query terms;   for each potential query term I:
 identifying, with the server, a subset of documents within the set of documents in which the potential query term I exists; 
   for each of the N−1 other potential query terms:
 analyzing, with the server, the subset of documents to identify a respective number of times the respective other potential query term occurs in the subset of documents; and 
 calculating, with the server, a co-occurrence score corresponding to matrix element (I,J) based on the respective number; and 
 constructing, with the server, the co-occurrence matrix using the co-occurrence scores. 
   
     
     
         2 . The method of  claim 1 , further comprising;
 receiving a query term;   identifying the row of the co-occurrence matrix corresponding to the query term;   retrieving at least non-zero co-occurrence scores stored in the identified row of the co-occurrence matrix.   
     
     
         3 . The method of  claim 1 , wherein constructing the co-occurrence matrix using the co-occurrence scores includes: storing only the co-occurrence scores greater than a threshold. 
     
     
         4 . The method of  claim 1 , wherein analyzing words includes:
 identifying which words correspond to nouns and verbs; and   using the nouns and verbs as the set of potential query terms.   
     
     
         5 . A non-transitory machine readable medium storing executable instructions which when executed by a system cause the system to perform a method of defining a data structure to use for expanding a search query requesting relevant software applications from a database storing software applications and a set of documents, each document being associated with a software application, the method comprising:
 identifying, with a server, a set of N potential query terms by analyzing words in the set of documents, N being an integer greater than 100;   defining a co-occurrence matrix defined by an N×N array of matrix elements, wherein each row corresponds to a respective one of the N potential query terms and each column corresponds to a respective one of the N potential query terms;   for each potential query term I:
 identifying, with the server, a subset of documents within the set of documents in which the potential query term I exists; 
   for each of the N−1 other potential query terms:
 analyzing, with the server, the subset of documents to identify a respective number of times the respective other potential query term J occurs in the subset of documents; and 
 calculating, with the server, a co-occurrence score corresponding to matrix element (I,J) based on the respective number; and 
 constructing, with the server, the co-occurrence matrix using the co-occurrence scores. 
   
     
     
         6 . The medium of  claim 5 , further comprising:
 receiving a query term;   identifying the row of the co-occurrence matrix corresponding to the query term;   retrieving at least non-zero co-occurrence scores stored in the identified row of the co-occurrence matrix.   
     
     
         7 . The medium of  claim 5 , wherein constructing the co-occurrence matrix using the co-occurrence scores includes: storing only the co-occurrence scores greater than a threshold. 
     
     
         8 . The medium of  claim 5 , wherein analyzing words includes:
 identifying which words correspond to nouns and verbs; and   using the nouns and verbs as the set of potential query terms.