Back to results
US 8,135,721 B2Granted
Discovering query intent from search queries and concept networks
AssigneeYahoo! Inc.
InventorsDeepa B. Joshi, John J. Thrall
Life of this patent
Estimating the term…
Abstract
A system is described for discovering query intent based on search queries and concept networks. The system may construct frequency vectors from log data corresponding to a submitted query and at least one related query submitted to one or more search engines. The system may also construct a query intent vector based on the frequency vectors. The query intent vector may include frequency scores that represent the intent of the query.
Claims19
A method of discovering the intent of a query comprising executing instructions stored on a memory that cause a computer processor to:
- analyze log data stored on a memory relating to a first query submitted by a user to one or more search engines;
- identify a plurality of related queries that are suggested to the user, where the identified related queries are related to the first query;
- analyze log data stored on the memory relating to the related queries;
- construct a first frequency vector for the first query and a related frequency vector for each of the related queries based on the analyzing, wherein the first frequency vector comprises an array of first query frequency score entries, each first query frequency score entry corresponding a different search engine in a set of search engines, wherein each related frequency vector comprises an array of related query frequency score entries, each related query frequency score entry corresponding to a different search engine in the set of search engines, and wherein each array has a length equal to the number of search engines in a set of search engines;
- determine a query intent vector comprising an array of entries, wherein the array has a length equal to the number of search engines in the seat of search engines, wherein each entry corresponds to a different search engine of the set of search engines, and wherein the intent of the query is derived from the query intent vector;
- compare each entry of the query intent vector to a threshold;
- for each entry of the query intent vector that exceeds the threshold, identify the search engine of the set of search engines that corresponds to that entry;
- and
- provide search results to the user based on a search of the first query on the identified search engines.
6 more elements
A product for discovering the intent of a query comprising:
- a memory;
- and
- instructions stored on the memory that, when executed, cause a computer processor to:
- analyze log data stored on the memory relating to a user query submitted by a user to one or more search engines;
- identify a plurality of related queries that are suggested to the user, where the identified related queries are related to the first query;
- analyze log data stored on the memory relating to the related queries;
- construct a first frequency vector for the user query and a related frequency vector for each of the related queries based on the analyzing, wherein the first frequency vector comprises an array of first query frequency score entries, each first query frequency score entry corresponding a different search engine in a set of search engines, wherein each related frequency vector comprises an array of related query frequency score entries, each related query frequency score entry corresponding to a different search engine in the set of search engines, and wherein each array has a length equal to the number of search engines in a set of search engines;
- determine a query intent vector comprising an array of entries, wherein the array has a length equal to the number of search engines in the seat of search engines, wherein each entry corresponds to a different search engine of the set of search engines, and wherein the intent of the user query is derived from the query intent vector;
- compare each entry of the query intent vector to a threshold;
- for each entry of the query intent vector that exceeds the threshold, identify the search engine of the set of search engines that corresponds to that entry;
- and
- provide search results to the user based on a search of the first query on the identified search engines.
9 more elements
A system for discovering the intent of a query comprising:
- a processor;
- and
- a memory coupled to the processor, the memory comprising:
- query log data;
- instructions stored on the memory that, when executed, cause the processor to:
- analyze query log data relating to a user query submitted by a user to one or more search engines;
- identify at least one related query that is suggested to the user;
- analyze query log data relating to the at least one related query;
- construct a first frequency vector for the user query and a second frequency vector for each of the at least one related query based on the analyzing, wherein the first frequency vector comprises an array of first query frequency score entries, each first query frequency score entry corresponding a different search engine in a set of search engines, wherein each second frequency vector comprises an array of related query frequency score entries, each related query frequency score entry corresponding to a different search engine in the set of search engines, and wherein each array has a length equal to the number of search engines in a set of search engines;
- determine a query intent vector based on an aggregate of a weighted first frequency vector and a weighted second frequency vector, wherein the query intent vector comprises an array of scores, wherein the array has a length equal to the number of search engines in the seat of search engines, wherein each score is associated with a different search engine of the set of search engines, and wherein the intent of the user query is derived from the query intent vector;
- compare each entry of the query intent vector to a threshold;
- for each entry of the query intent vector that exceeds the threshold, identify the search engine of the set of search engines that corresponds to that entry;
- and
- provide search results to the user based on a search of the first query on the identified search engines.
11 more elements
Dependencies are inferred from the claim text.