Methods, systems, and computer program products for integrated world wide web query classification

Patent metadata

Field	Value
Publication number	US-9465862-B2
Application number	US-201514614606-A
Country	US
Kind code	B2
Filing date	Feb 5, 2015
Priority date	Jul 26, 2012
Publication date	Oct 11, 2016
Grant date	Oct 11, 2016

Abstract

Implementing query classification includes executing a reductionist module on a query to extract a core term, which term is used to search a hash table that maps core terms to corresponding categories, deriving a first result including one of the categories from the search, and executing an enrichment module on the query to yield a second result. The enrichment module includes searching an index of terms that are mapped to documents and corresponding categories. Upon determining the core term is present in the hash table, a weighted average is calculated for values of the first and second results based on training data. Upon determining the core term from the query is not in the hash table, and also that a probability score of the category in the index for the second result meets a minimum confidence value, the core term and the corresponding categories are stored in the hash table.

First claim

What is claimed is:

1. A method for integrating query categories, comprising:
executing, at a computer, a reductionist module on a search query to extract a core term from the search query, the core term used to search a hash table that maps core terms to corresponding categories;
deriving a first result comprising at least one of the categories from the search of the hash table;
executing at the computer an enrichment module on the search query to yield a second result, the enrichment module including searching an index of terms that are mapped to documents and corresponding categories in the index, the second result indicative of one of the corresponding categories in the index based on a probability score;
upon determining the core term is present in the hash table, calculating a weighted average for corresponding values of the first result and the second result based on training data acquired from the execution of the reductionist module and the execution of the enrichment module, the calculated weighted average stored in a memory device; and
upon determining the core term from the search query is not listed in the hash table, and upon determining the probability score of the one of the corresponding categories in the index for the second result meets a minimum defined confidence value, inserting and storing the core term and the one of the corresponding categories in the hash table and mapping the core term to the one of the corresponding categories in the hash table.

2. The method of claim 1 , wherein executing the reductionist module includes parsing the search query via a natural language parser.

3. The method of claim 1 , further comprising:
providing the core term to the enrichment module, the enrichment module associating a minimum weight to the core term in the execution of the enrichment module.

4. The method of claim 1 , further comprising:
upon determining the core term from the search query is not listed in the hash table, providing a synonym of the core term to the enrichment module, the enrichment module expanding the search query to include the synonym.

5. The method of claim 1 , wherein the weighted average is applied equally to the first result and the second result based on training data derived from the execution of the reductionist module and the execution of the enrichment module.

6. The method of claim 1 , wherein the calculating yields a third result, the method further comprising:
transmitting the third result to a computer device that generated the search query.

7. The method of claim 1 , wherein the reductionist module is executed simultaneously with the enrichment module.

8. A system for integrating query categories, comprising:
a computer processor; and
logic executable by the computer processor, the logic implementing a method, the method including:
executing a reductionist module on a search query to extract a core term from the search query, the core term used to search a hash table that maps core terms to corresponding categories;
deriving a first result comprising at least one of the categories from the search of the hash table;
executing an enrichment module on the search query to yield a second result, the enrichment module including searching an index of terms that are mapped to documents and corresponding categories in the index, the second result indicative of one of the corresponding categories in the index based on a probability score;
upon determining the core term is present in the hash table, calculating a weighted average for corresponding values of the first result and the second result based on training data acquired from the execution of the reductionist module and the execution of the enrichment module, the calculated weighted average stored in a memory device; and
upon determining the core term from the search query is not listed in the hash table, and upon determining the probability score of the one of the corresponding categories in the index for the second result meets a minimum defined confidence value, inserting and storing the core term and the one of the corresponding categories in the hash table and mapping the core term to the one of the corresponding categories in the hash table.

9. The system of claim 8 , wherein executing the reductionist module includes parsing the search query via a natural language parser.

10. The system of claim 8 , wherein the logic further implements:
providing the core term to the enrichment module, the enrichment module associating a minimum weight to the core term in the execution of the enrichment module.

11. The system of claim 8 , wherein the logic further implements:
upon determining the core term from the search query is not listed in the table, providing a synonym of the core term to the enrichment module, the enrichment module expanding the search query to include the synonym.

12. The system of claim 8 , wherein the weighted average is applied equally to the first result and the second result based on training data derived from the execution of the reductionist module and the execution of the enrichment module.

13. The system of claim 8 , wherein the calculating yields a third result, the logic further implementing:
transmitting the third result to a computer device that generated the search query.

14. The system of claim 8 , wherein the reductionist module is executed simultaneously with the enrichment module.

15. A computer program product embodied on a non-transitory computer readable storage medium, the computer program product including instructions for causing a computer to execute a method for integrating query categories, comprising:
executing a reductionist module on the search query to extract a core term from the search query, the core term used to search a hash table that maps core terms to corresponding categories;
deriving a first result comprising at least one of the categories from the search of the hash table;
executing an enrichment module on the search query to yield a second result, the enrichment module including searching an index of terms that are mapped to documents and corresponding categories in the index, the second result indicative of one of the corresponding categories in the index based on a probability score;
upon determining the core term is present in the hash table, calculating a weighted average for corresponding values of the first result and the second result based on training data acquired from the execution of the reductionist module and the execution of the enrichment module, the calculated weighted average stored in a memory device; and
upon determining the core term from the search query is not listed in the hash table, and upon determining the probability score of the one of the corresponding categories in the index for the second result meets a minimum defined confidence value, inserting and storing the core term and the one of the corresponding categories in the hash table and mapping the core term to the one of the corresponding categories in the hash table.

16. The computer program product of claim 15 , wherein executing the reductionist module includes parsing the search query via a natural language parser.

17. The computer program product of claim 15 , wherein the method further includes:
providing the core term to the enrichment module, the enrichment module associating a minimum weight to the core term in the execution of the enrichment module.

18. The computer program product of claim 15 , wherein the method further includes:
upon determining the core term from the search query i

Assignees

• At & T Ip I Lp

Inventors

• Agrawal Ritesh Jitendra
• King Irwin
• Zajac Remi

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06F17/30654

  Physics · mapped topic

• G06F17/30734

  Physics · mapped topic

• G06F17/30616

  Physics · mapped topic

• G06F17/30864

  Physics · mapped topic