Classifying queries

• US 9,229,974 B1
• Filed: 10/22/2012
• Issued: 01/05/2016
• Est. Priority Date: 06/01/2012
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Lee, John J., Carroll, Melissa K.
• Primary Examiner(s): Hoang, Son T
• Application Number:
  US13/656,959
• Time in Patent Office: 1,170 Days
• Field of Search: 707/722
• US Class Current: 1/1
• CPC Class Codes: G06F 16/245 Query processing
• CPC Class Codes: G06F 16/35 Clustering; Classification

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for classifying queries. A query that includes one or more query terms is received. One or more entity identifiers are extracted from one or more search results that are responsive to the query. One or more of the query terms are compared with the extracted entity identifiers. Based on comparing the query terms with the extracted entity identifiers, the query is classified as an entity-triggering query or as a description-triggering query.

Claims

1. A method comprising:

receiving, by a computer processor, a query that includes one or more query terms;

extracting, by the computer processor, one or more entity identifiers from one or more search results that are responsive to the query;

determining, by the computer processor, whether all of the extracted entity identifiers match or do not match at least one of the one or more query terms; and

classifying the query as a description-triggering query or as an entity-triggering query, the query being classified as a description-triggering query based on determining that all of the extracted entity identifiers match the at least one of the one or more query terms, or the query being classified as an entity-triggering query based on determining that all of the extracted entity identifiers do not match the at least one of the one or more query terms;

wherein a description triggering query is a query for which a natural language snippet associated with one or more of the search results is provided as an answer to the query, and wherein an entity-triggering query is a query for which an entity identifier that does not match one or more of the query terms is provided as an answer to the query.

2. The method of claim 1, comprising:

based on classifying the query as the entity-triggering query, selecting a response format designated for queries classified as entity-triggering queries, wherein the selected response format is not designated for queries classified as description-triggering queries; and

providing a response to the query using the selected response format.

3. The method of claim 1, comprising:

based on classifying the query as the description-triggering query, selecting a response format designated for queries classified as description-triggering queries, wherein the selected response format is not designated for queries classified as entity-triggering queries; and

providing a response to the query using the selected response format.

4. The method of claim 1, wherein:

extracting one or more entity identifiers from the one or more search results comprises extracting multiple entity identifiers from the one or more search results.

5. The method of claim 4, comprising:

determining that a plurality of entity identifiers in the extracted multiple entity identifiers do not include any of the one or more query terms;

selecting, from among the plurality of entity identifiers, an entity identifier that was extracted from a highest-scoring search result of the identified one or more search results; and

indicating that the selected entity identifier is the answer to the query.

6. The method of claim 4, comprising:

determining that a plurality of entity identifiers in the extracted multiple entity identifiers do not include any of the one or more query terms;

counting, for each entity identifier in the plurality of entity identifiers, a number of the one or more search results from which the entity identifier was extracted;

selecting, from among the plurality of entity identifiers, the entity identifier that was extracted from the highest number of the one or more search results; and

indicating that the selected entity identifier is the answer to the query.

7. The method of claim 1, wherein extracting one or more entity identifiers from the one or more search results comprises, for at least one of the one or more search results:

identifying a title of a document associated with the at least one search result;

identifying a title format associated with the document; and

extracting, using the title format, a portion of the title as the entity identifier associated with the at least one search result.

8. The method of claim 7, wherein identifying the title format associated with the document includes identifying an Internet domain associated with the document, the title format being associated with the Internet domain.

9. The method of claim 1, wherein:

identifying the one or more search results that are responsive to the query comprises identifying one or more search results that are each associated with one of a plurality of different information sources;

the method further comprises identifying a proper subset of the one or more search results, the proper subset comprising only one or more search results that are associated with a particular information source; and

extracting the one or more entity identifiers from the search results comprises extracting the one or more entity identifiers from only the proper subset of the one or more search results.

10. The method of claim 9, wherein identifying the proper subset of the one or more search results comprises identifying one or more search results that reference a resource hosted in a particular Internet domain.

11. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising;

receiving a query that includes one or more query terms;

extracting one or more entity identifiers from one or more search results that are responsive to the query;

determining whether all of the extracted entity identifiers match or do not match at least one of the one or more query terms; and

classifying the query as a description-triggering query or as an entity-triggering query, the query being classified as a description-triggering query based on determining that all of the extracted entity identifiers match the at least one of the one or more query terms, or the query being classified as an entity-triggering query based on determining that all of the extracted entity identifiers do not match the at least one of the one or more query terms;

wherein a description triggering query is a query for which a natural language snippet associated with one or more of the search results is provided as an answer to the query, and wherein an entity-triggering query is a query for which an entity identifier that does not match one or more of the query terms is provided as an answer to the query.

12. The system of claim 11, wherein the operations comprise:

based on classifying the query as the entity-triggering query, selecting a response format designated for queries classified as entity-triggering queries, wherein the selected response format is not designated for queries classified as description-triggering queries; and

providing a response to the query using the selected response format.

13. The system of claim 11, wherein the operations comprise:

based on classifying the query as the description-triggering query, selecting a response format designated for queries classified as description-triggering queries, wherein the selected response format is not designated for queries classified as entity-triggering queries; and

providing a response to the query using the selected response format.

14. The system of claim 11, wherein:

extracting one or more entity identifiers from the one or more search results comprises extracting multiple entity identifiers from the one or more search results.

15. The system of claim 14, wherein the operations comprise:

determining that a plurality of entity identifiers in the extracted multiple entity identifiers do not include any of the one or more query terms;

selecting, from among the plurality of entity identifiers, an entity identifier that was extracted from a highest-scoring search result of the identified one or more search results; and

indicating that the selected entity identifier is the answer to the query.

16. The system of claim 14, wherein the operations comprise:

determining that a plurality of entity identifiers in the extracted multiple entity identifiers do not include any of the one or more query terms;

counting, for each entity identifier in the plurality of entity identifiers, a number of the one or more search results from which the entity identifier was extracted;

selecting, from among the plurality of entity identifiers, the entity identifier that was extracted from the highest number of the one or more search results; and

indicating that the selected entity identifier is the answer to the query.

17. The system of claim 11, wherein extracting one or more entity identifiers from the one or more search results comprises, for at least one of the one or more search results:

identifying a title of a document associated with the at least one search result;

identifying a title format associated with the document; and

extracting, using the title format, a portion of the title as the entity identifier associated with the at least one search result.

18. The system of claim 17, wherein identifying the title format associated with the document includes identifying an Internet domain associated with the document, the title format being associated with the Internet domain.

19. A non-transitory computer storage medium encoded with a computer program, the program comprising instructions that when executed by one or more computers cause the one or more computers to perform operations comprising:

receiving a query that includes one or more query terms;

extracting one or more entity identifiers from one or more search results that are responsive to the query;

determining whether all of the extracted entity identifiers match or do not match at least one of the one or more query terms; and

classifying the query as a description-triggering query or as an entity-triggering query, the query being classified as a description-triggering query based on determining that all of the extracted entity identifiers match the at least one of the one or more query terms, or the query being classified as an entity-triggering query based on determining that all of the extracted entity identifiers do not match the at least one of the one or more query terms;

wherein a description triggering query is a query for which a natural language snippet associated with one or more of the search results is provided as an answer to the query, and wherein an entity-triggering query is a query for which an entity identifier that does not match one or more of the query terms is provided as an answer to the query.