Factual query pattern learning

• US 9,898,512 B1
• Filed: 05/13/2015
• Issued: 02/20/2018
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Xian, Junli, Sahin, Engin Cinar, Blitzer, John, Persky, Emma S.
• Primary Examiner(s): Stevens, Robert
• Application Number:
  US14/710,794
• Time in Patent Office: 1,014 Days
• Field of Search: 707727
• CPC Class Codes: G06F 16/14 Details of searching files based on file metadata
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/90332 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing a statement that references a particular attribute of a particular topic, in response to providing the statement, obtaining one or more query patterns that each include one or more query terms that are used in queries submitted to a search system in obtaining a value for the particular attribute of the particular topic, generalizing one or more of the query patterns, and associating the one or more generalized query patterns with one or more other topics that include the particular attribute.

Claims

1. A computer-implemented method comprising:

obtaining a collection of query patterns that each include one or more query terms that one or more users of a search engine has indicated are useful in eliciting the search engine to provide an answer that indicates a value for a particular attribute of a particular topic;

selecting, from among the collection of query patterns that each include one or more query terms that the one or more users of the search engine has indicated are useful in eliciting the search engine to provide the answer that indicates the value for the particular attribute of the particular topic, a subset of the candidate query patterns of the collection;

generalizing one or more of the query patterns of the subset to generate one or more generalized query patterns for eliciting the search engine to provide an answer that indicates a value for the particular attribute of other topics that are different than the particular topic;

associating the one or more generalized query patterns with one or more of the other topics that include the particular attribute;

receiving a search query that includes the particular attribute and a topic that is (i) different than the particular topic, and (ii) is associated with a particular generalized query pattern from among the one or more generalized query patterns;

retrieving a particular value for the particular attribute of the received search query, the particular value being specified by the particular generalized query pattern that includes the topic of the received search query; and

generating an answer to the search query that includes the retrieved value for the particular value for the particular attribute of the received search query.

2. The method of claim 1, comprising:

receiving a query from a user;

determining that the query includes a first portion that matches a particular, generalized query pattern; and

based on determining that the query includes the first portion that matches the particular, generalized query pattern, retrieving a value of the particular attribute of one of the other topics referenced by the query.

3. The method of claim 1, wherein obtaining one or more query patterns comprises:

receiving query patterns relating to a statement; and

normalizing the received query patterns.

4. The method of claim 3, comprising:

filtering the received query patterns to remove query patterns that occur with a frequency that does not satisfy a predetermined threshold.

5. The method of claim 1, wherein generalizing one or more of the query patterns comprises replacing a first portion of each of the one or more query patterns that match the particular topic with one or more identifiers.

6. The method of claim 1, wherein the query patterns are received from a plurality of human users.

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

obtaining a collection of query patterns that each include one or more query terms that one or more users of a search engine has indicated are useful in eliciting the search engine to provide an answer that indicates a value for a particular attribute of a particular topic;

selecting, from among the collection of query patterns that each include one or more query terms that the one or more users of the search engine has indicated are useful in eliciting the search engine to provide the answer that indicates the value for the particular attribute of the particular topic, a subset of the candidate query patterns of the collection;

generalizing one or more of the query patterns of the subset to generate one or more generalized query patterns for eliciting the search engine to provide an answer that indicates a value for the particular attribute of other topics that are different than the particular topic;

associating the one or more generalized query patterns with one or more of the other topics that include the particular attribute;

receiving a search query that includes the particular attribute and a topic that is (i) different than the particular topic, and (ii) is associated with a particular generalized query pattern from among the one or more generalized query patterns;

retrieving a particular value for the particular attribute of the received search query, the particular value being specified by the particular generalized query pattern that includes the topic of the received search query; and

generating an answer to the search query that includes the retrieved value for the particular value for the particular attribute of the received search query.

8. The device of claim 7, wherein the operations comprise:

receiving a query from a user;

determining that the query includes a first portion that matches a particular, generalized query pattern; and

based on determining that the query includes the first portion that matches the particular, generalized query pattern, retrieving a value of the particular attribute of one of the other topics referenced by the query.

9. The device of claim 7, wherein obtaining one or more query patterns comprises:

receiving query patterns relating to a statement; and

normalizing the received query patterns.

10. The device of claim 9, wherein the operations comprise:

filtering the received query patterns to remove query patterns that occur with a frequency that does not satisfy a predetermined threshold.

11. The device of claim 9, wherein the statement is selected to cause one or more query patterns to be obtained that include the particular attribute for which associations for the one or more generalized query patterns that include the particular attribute are needed.

12. The device of claim 7, wherein generalizing one or more of the query patterns comprises replacing a first portion of each of the one or more query patterns that match the particular topic with one or more identifiers.

13. 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;

obtaining a collection of query patterns that each include one or more query terms that one or more users of a search engine has indicated are useful in eliciting the search engine to provide an answer that indicates a value for a particular attribute of a particular topic;

selecting, from among the collection of query patterns that each include one or more query terms that the one or more users of the search engine has indicated are useful in eliciting the search engine to provide the answer that indicates the value for the particular attribute of the particular topic, a subset of the candidate query patterns of the collection;

generalizing one or more of the query patterns of the subset to generate one or more generalized query patterns for eliciting the search engine to provide an answer that indicates a value for the particular attribute of other topics that are different than the particular topic;

associating the one or more generalized query patterns with one or more of the other topics that include the particular attribute;

receiving a search query that includes the particular attribute and a topic that is (i) different than the particular topic, and (ii) is associated with a particular generalized query pattern from among the one or more generalized query patterns;

retrieving a particular value for the particular attribute of the received search query, the particular value being specified by the particular generalized query pattern that includes the topic of the received search query; and

generating an answer to the search query that includes the retrieved value for the particular value for the particular attribute of the received search query.

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

receiving a query from a user;

determining that the query includes a first portion that matches a particular, generalized query pattern; and

based on determining that the query includes the first portion that matches the particular, generalized query pattern, retrieving a value of the particular attribute of one of the other topics referenced by the query.

15. The system of claim 13, wherein obtaining one or more query patterns comprises:

receiving query patterns relating to a statement; and

normalizing the received query patterns.

16. The system of claim 15, wherein the statement is selected to cause one or more query patterns to be obtained that include the particular attribute for which associations for the one or more generalized query patterns that include the particular attribute are needed.

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

filtering the received query patterns to remove query patterns that occur with a frequency that does not satisfy a predetermined threshold.

18. The system of claim 13, wherein generalizing one or more of the query patterns comprises replacing a first portion of each of the one or more query patterns that match the particular topic with one or more identifiers.