Disambiguating associations

Patent metadata

Field	Value
Publication number	US-8949264-B2
Application number	US-201213361326-A
Country	US
Kind code	B2
Filing date	Jan 30, 2012
Priority date	Jan 30, 2012
Publication date	Feb 3, 2015
Grant date	Feb 3, 2015

Abstract

Disclosed herein is a technique for disambiguating associations between one keyword and multiple attributes of a database model and for disambiguating associations between one attribute of a database model and multiple attribute types.

First claim

The invention claimed is:

1. A system comprising:
at least one processor to:
access associations between keywords contained in formerly received natural language queries and attributes of a database model;
access a context free grammar comprising a first sentence that includes elements of a formerly received natural language query, wherein in the first sentence a given keyword in the formerly received natural language query that is associated with an attribute of the database model is replaced with a first label representing a first one of plural different database attribute types of the attributes of the database model;
disambiguate first associations between a first keyword and multiple attributes of the database model;
disambiguate second associations between an attribute of the database model and the plural different database attribute types using the context free grammar; and
respond to a new natural language query by generating a database language query based on the disambiguating of the first and second associations.

2. The system of claim 1 , wherein the context free grammar further includes a second sentence including the elements of the formerly received natural language query, the second sentence including a second label that replaces the given keyword in the formerly received natural language query, the second label representing a second, different one of the plural different database attribute types.

3. The system of claim 2 , wherein generating the database language query is further based on rankings associated with the first and second sentences, the rankings corresponding to likelihoods that respective ones of the first and second sentences will translate into respective expressions of the database language query that produce a correct answer to the new natural language query.

4. A system comprising:
at least one processor to:
learn to translate natural language queries into database language queries using historical data associated with previously received natural language queries;
respond to a new natural language query based on an analysis of the historical data, the historical data comprising:
associations between keywords contained in the previously received natural language queries and attributes of a database model, the associations including associations between a first of the keywords and multiple attributes of the database model; and
a context free grammar, the context free grammar being adaptable for disambiguating associations between an attribute of the database model and multiple attribute types, the multiple attribute types comprising a database table, a database column, and a value in a database table, the context free grammar further comprising a sentence including elements of a formerly received natural language query, and including a label replacing a given keyword in the formerly received natural language query and representing one of the database attribute types, the one database attribute type corresponding to the attribute of the database model associated with the given keyword; and

associate a given association between a keyword in the new natural language query and a respective attribute of the database model with a first probability, the first probability representing a likelihood that the given association will translate into at least one expression of a database language query that produces a correct answer to the new natural language query.

5. The system of claim 4 , wherein the at least one processor is to associate the sentence with a second probability, the second probability representing a likelihood that the sentence will translate into at least one expression of the database language query that produces the correct answer to the new natural language query.

6. The system of claim 5 , wherein the at least one processor is to further assign a first ranking score to the given association between the keyword in the new natural language query and the respective attribute of the database model based on the first probability.

7. The system of claim 6 , wherein the at least one processor is to further assign a second ranking score to the sentence based on the second probability.

8. The system of claim 7 , wherein the at least one processor is to further:
generate a final ranking score for a combination of first ranking score and the second ranking score;
generate at least one database language query for the new natural language query such that expressions of the at least one database language query include attributes of the database model and attribute types associated with the database model that correspond to the final ranking score; and
execute the at least one database language query.

9. A non-transitory computer readable medium having instructions stored therein for causing at least one processor to:
access associations between keywords contained in formerly received natural language queries and attributes of a database model;
access a context free grammar comprising a first sentence that includes elements of a formerly received natural language query, wherein in the first sentence a given keyword in the formerly received natural language query that is associated with an attribute of the database model is replaced with a first label representing a first one of plural different database attribute types of the attributes of the database model;
disambiguate first associations between a first keyword and multiple attributes of the database model;
disambiguate second associations between an attribute of the database model and the plural different database attribute types using the context free grammar; and
respond to a new natural language query by generating a database language query based on the disambiguating of the first and second associations.

10. The non-transitory computer readable medium of claim 9 , wherein the instructions are for further causing the at least one processor to assign a first probability to a given association between a keyword in the new natural language query and a respective attribute of the database model, the first probability representing a likelihood that the given association will translate into at least one expression of the database language query that produces a correct answer to the new natural language query.

11. The non-transitory computer readable medium of claim 10 , wherein the instructions are for further causing the at least one processor to assign a second probability to the sentence of the context free grammar, the second probability representing the likelihood that the sentence will translate into at least one expression of the database language query that produces the correct answer to the new natural language query.

12. The non-transitory computer readable medium of claim 11 , wherein the instructions are for further causing the at least one processor to assign a first ranking score to the given association between the keyword in the new natural language query and the respective attribute of the database model based on the first probability.

13. The non-transitory computer readable medium of claim 12 , wherein the instructions are for further causing the at least one processor to assign a second ranking score to the sentence based on the second probability.

14. The non-transitory computer readable medium of claim 13 , wherein the instructions are for further causing the at least one processor to:
generate a final ranking score for a combination of the first ranking score and the second ranking score;
generate the database language query such that expressions of the database language query include attributes of the database model a

Assignees

• Cohen Ira
• Assulin Ohad
• Hewlett Packard Development Co

Inventors

• Cohen Ira
• Assulin Ohad

Classifications

• G06F16/243 Primary

  Natural language query formulation · CPC title