Detecting name-triggering queries

• US 9,063,983 B1
• Filed: 03/12/2013
• Issued: 06/23/2015
• 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.
• Primary Examiner(s): VO, TRUONG V
• Application Number:
  US13/795,650
• Time in Patent Office: 833 Days
• Field of Search: 707/706
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24565 Triggers; Constraints
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

A system may be configured to receive a query; identify an object and a verb associated with the query; obtain information indicating a quantity of occurrences, in one or more documents, of the object in a sentence that may include the verb, or a quantity of occurrences, in one or more documents, of one or more terms, which are related to the object, in a sentence that may include the verb; generate a confidence score that may indicate a probability that the query seeks a result relating to a name of a person; identify a set of documents that are responsive to the query; determine that the confidence score satisfies a threshold; identify one or more documents, of the set of documents, that are associated with one or more names of people; select a particular document, of the set of documents; and output information regarding the selected particular document.

Claims

1. A computer-implemented method comprising:

receiving a search query;

determining that the search query is classified as a potential name-triggering query, wherein a name-triggering query is a query for which a response is to include a person'"'"'s name;

based on determining that the search query is classified as a potential name-triggering query, extracting a verb and an object from the search query;

identifying a set of data structures, from among a repository of data structures, that store data referencing a form of the verb that was extracted from the search query;

identifying a subset of the data structures that store data referencing the object, or a term that is related to the object;

determining a name-triggering score for the search query based on the subset of data structures, wherein the count is determined based on a quantity of occurrences of the object being associated, in a collection of resources, with a person'"'"'s name;

identifying a set of resources that are responsive to the search query;

determining that the name-triggering score satisfies a threshold;

identifying, based on determining that the name-triggering score satisfies the threshold, one or more resources, of the set of resources, that are associated with one or more names of one or more people;

selecting a particular resource from the set of resources; and

outputting information regarding the selected particular resource.

2. The method of claim 1, wherein determining a name-triggering score for the search query based on one or more counts identified in the subset of data structures comprises:

determining a first score based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing the object and based on a quantity of the data structures that (i) store data referencing a form of the verb and (ii) store data referencing a term that is related to the object;

determining a second score based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing the object and not based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing a term that is related to the object;

determining a third score not based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing the object and based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing a term that is related to the object; and

combining the first score, the second score, and the third score to determine the name-triggering score.

3. The method of claim 2, wherein combining the first score, the second score, and the third score to determine the name-triggering score comprises:

calculating a weighted average of the first score, the second score, and the third score, wherein at least two of the first score, the second score, and the third score are associated with different weights.

4. The method of claim 3, wherein the first score is associated with a first weight, the second score is associated with a second weight, and the third score is associated with a third weight,

wherein the second weight is greater than the first weight, and

wherein the first weight is greater than the third weight.

5. The method of claim 1, comprising:

identifying a resource score associated with each of the one or more resources that are associated with one or more names of one or more people; and

adjusting the resource score associated with each of the one or more resources that are associated with one or more names of one or more people,

wherein selecting a particular resource from the set of resources is based on the adjusted resource scores.

6. The method of claim 5, wherein determining that the name-triggering score satisfies the threshold comprises:

determining that a response to the search query includes a name of a person,

wherein adjusting the resource score associated with each of the one or more resources that are associated with one or more names of one or more people includes increasing the resource score associated with each of the one or more resources that are associated with one or more names of one or more people, and

wherein selecting a particular resource from the set of resources is based on the adjusted scores comprises;

determining that the particular resource from the set of resources is associated with a highest resource score of the resource scores associated with the set of resources.

7. The method of claim 6, comprising:

decreasing the resource score associated with each of one or more resources that are not identified as being associated with one or more names of one or more people.

8. The method of claim 5, wherein determining that the name-triggering score satisfies the threshold comprises:

determining that a response to the search query does not include a name of a person,

wherein adjusting the resource score associated with each of the one or more resources that are associated with one or more names of one or more people includes increasing the resource score associated with each of the one or more resources that are associated with one or more names of one or more people, and

wherein selecting a particular resource from the set of resources is based on the adjusted scores comprises;

determining that the particular resource from the set of resources is associated with a highest resource score of the resource scores associated with the set of resources.

9. 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 search query;

determining that the search query is classified as a potential name-triggering query, wherein a name-triggering query is a query for which a response is to include a person'"'"'s name;

based on determining that the search query is classified as a potential name-triggering query, extracting a verb and an object from the search query;

identifying a set of data structures, from among a repository of data structures, that store data referencing a form of the verb that was extracted from the search query;

identifying a subset of the data structures that store data referencing the object, or a term that is related to the object;

determining a name-triggering score for the search query based on the subset of data structures, wherein the count is determined based on a quantity of occurrences of the object being associated, in a collection of resources, with a person'"'"'s name;

identifying a set of resources that are responsive to the search query;

determining that the name-triggering score satisfies a threshold;

identifying, based on determining that the name-triggering score satisfies the threshold, one or more resources, of the set of resources, that are associated with one or more names of one or more people;

selecting a particular resource from the set of resources; and

outputting information regarding the selected particular resource.

10. The system of claim 9, wherein determining a name-triggering score for the search query based on the subset of data structures comprises:

determining a first score based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing the object and based on a quantity of the data structures that (i) store data referencing a form of the verb and (ii) store data referencing a term that is related to the object;

determining a second score based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing the object and not based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing a term that is related to the object;

determining a third score not based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing the object and based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing a term that is related to the object; and

combining the first score, the second score, and the third score to determine the name-triggering score.

11. The system of claim 10, wherein combining the first score, the second score, and the third score to determine the name-triggering score comprises:

calculating a weighted average of the first score, the second score, and the third score, wherein at least two of the first score, the second score, and the third score are associated with different weights.

12. The system of claim 11, wherein the first score is associated with a first weight, the second score is associated with a second weight, and the third score is associated with a third weight,

wherein the second weight is greater than the first weight, and

wherein the first weight is greater than the third weight.

13. The system of claim 9, comprising:

identifying a resource score associated with each of the one or more resources that are associated with one or more names of one or more people; and

adjusting the resource score associated with each of the one or more resources that are associated with one or more names of one or more people,

wherein selecting a particular resource from the set of resources is based on the adjusted resource scores.

14. The system of claim 13, wherein determining that the name-triggering score satisfies the threshold comprises:

determining that a response to the search query includes a name of a person,

wherein adjusting the resource score associated with each of the one or more resources that are associated with one or more names of one or more people includes increasing the resource score associated with each of the one or more resources that are associated with one or more names of one or more people, and

wherein selecting a particular resource from the set of resources is based on the adjusted scores comprises;

determining that the particular resource from the set of resources is associated with a highest resource score of the resource scores associated with the set of resources.

15. The system of claim 14, comprising:

decreasing the resource score associated with each of one or more resources that are not identified as being associated with one or more names of one or more people.

16. The system of claim 13, wherein determining that the name-triggering score satisfies the threshold comprises:

determining that a response to the search query does not include a name of a person,

wherein adjusting the resource score associated with each of the one or more resources that are associated with one or more names of one or more people includes increasing the resource score associated with each of the one or more resources that are associated with one or more names of one or more people, and

wherein selecting a particular resource from the set of resources is based on the adjusted scores comprises;

determining that the particular resource from the set of resources is associated with a highest resource score of the resource scores associated with the set of resources.

17. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:

receiving a search query;

determining that the search query is classified as a potential name-triggering query, wherein a name-triggering query is a query for which a response is to include a person'"'"'s name;

based on determining that the search query is classified as a potential name-triggering query, extracting a verb and an object from the search query;

identifying a set of data structures, from among a repository of data structures, that store data referencing a form of the verb that was extracted from the search query;

identifying a subset of the data structures that store data referencing the object, or a term that is related to the object;

determining a name-triggering score for the search query based on the subset of data structures, wherein the count is determined based on a quantity of occurrences of the object being associated, in a collection of resources, with a person'"'"'s name;

identifying a set of resources that are responsive to the search query;

determining that the name-triggering score satisfies a threshold;

identifying, based on determining that the name-triggering score satisfies the threshold, one or more resources, of the set of resources, that are associated with one or more names of one or more people;

selecting a particular resource from the set of resources; and

outputting information regarding the selected particular resource.

18. The medium of claim 17, wherein determining a name-triggering score for the search query based on the subset of data structures comprises:

determining a first score based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing the object and based on a quantity of the data structures that (i) store data referencing a form of the verb and (ii) store data referencing a term that is related to the object;

determining a second score based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing the object and not based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing a term that is related to the object;

determining a third score not based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing the object and based on a quantity of the data structures that (i) store data referencing a form of the verb that was extracted from the search query and (ii) store data referencing a term that is related to the object; and

combining the first score, the second score, and the third score to determine the name-triggering score.

19. The medium of claim 17, comprising:

identifying a resource score associated with each of the one or more resources that are associated with one or more names of one or more people; and

adjusting the resource score associated with each of the one or more resources that are associated with one or more names of one or more people,

wherein selecting a particular resource from the set of resources is based on the adjusted resource scores.

20. The medium of claim 19, wherein determining that the name-triggering score satisfies the threshold comprises:

determining that a response to the search query includes a name of a person,

wherein adjusting the resource score associated with each of the one or more resources that are associated with one or more names of one or more people includes increasing the resource score associated with each of the one or more resources that are associated with one or more names of one or more people, and

wherein selecting a particular resource from the set of resources is based on the adjusted scores comprises;

determining that the particular resource from the set of resources is associated with a highest resource score of the resource scores associated with the set of resources.