Systems and methods to facilitate local searches via location disambiguation

• US 9,424,529 B2
• Filed: 11/24/2014
• Issued: 08/23/2016
• Est. Priority Date: 11/04/2010
• Status: Active Grant

• Current Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Agrawal, Ritesh Jitendra, Shanahan, James G.
• Primary Examiner(s): Hill, Stanley K
• Assistant Examiner(s): Misir, Dave
• Application Number:
  US14/551,344
• Publication Number:
  US 20150081603A1
• Time in Patent Office: 638 Days
• Field of Search: 706/12, 707/711, 707/E17.108
• US Class Current: 1/1
• CPC Class Codes: G06F 16/29 Geographical information databases
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9537 Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
• CPC Class Codes: G06N 20/00 Machine learning

Abstract

Systems and methods use machine learning techniques to resolve location ambiguity in search queries. In one aspect, a dataset generator generates a training dataset using query logs of a search engine. A training engine applies a machine learning technique to the training dataset to generate a location disambiguation model. A location disambiguation engine uses the location disambiguation model to resolve location ambiguity in subsequent search queries.

Claims

1. A method comprising:

filtering, by a system comprising a processor, a query log to identify a first query, the first query specifying an unambiguous location;

generating, by the system from the first query, a second query having location ambiguity, wherein generating, from the first query, the second query having location ambiguity comprises

separating, by the system, the first query into at least a first component and a second component, and

removing, by the system, one of the first component or the second component;

identifying, by the system, a set of location candidates for the second query;

generating, by the system, a training dataset based on comparing the unambiguous location specified in the first query with each location candidate of the set of location candidates for the second query;

applying, by the system, a machine learning technique to the training dataset to generate a location disambiguation model; and

resolving, by the system, location ambiguity in a query using the location disambiguation model.

2. The method of claim 1, wherein the machine learning technique comprises a decision tree learning technique.

3. The method of claim 1, wherein the machine learning technique comprises a gradient boosted decision tree learning technique.

4. The method of claim 1, wherein the training dataset further comprises a location feature of a location candidate of the set of location candidates and a search feature associated with the location candidate of the set of location candidates.

5. The method of claim 4, wherein the location feature is independent of queries recorded in the query log.

6. The method of claim 4, wherein the location feature comprises at least one of a population of the location candidate, a population density of the location candidate, demography of the location candidate, a popularity of the location candidate, or a business population of the location candidate.

7. The method of claim 4, wherein the search feature is dependent on queries recorded in the query log and independent of searchers of the queries recorded in the query log.

8. The method of claim 1, wherein applying the machine learning technique to the training dataset to generate the location disambiguation model comprises generating a first location model for queries received from mobile devices and generating a second location model for queries received from fixed-location devices.

9. A non-transitory computer readable medium storing instructions which, when executed by a system comprising a processor, cause the processor to perform operations comprising:

filtering a query log to identify a first query, the first query specifying an unambiguous location;

generating, from the first query, a second query having location ambiguity, wherein generating, from the first query, the second query having location ambiguity comprises

separating the first query into at least a first component and a second component, and

removing one of the first component or the second component;

identifying a set of location candidates for the second query;

generating a training dataset based on comparing the unambiguous location specified in the first query with each location candidate of the set of location candidates for the second query;

applying a machine learning technique to the training dataset to generate a location disambiguation model; and

resolving location ambiguity in a query using the location disambiguation model.

10. The non-transitory computer readable medium of claim 9, wherein the machine learning technique comprises a decision tree learning technique.

11. The non-transitory computer readable medium of claim 9, wherein the machine learning technique comprises a gradient boosted decision tree learning technique.

12. The non-transitory computer readable medium of claim 9, wherein the training dataset further comprises a location feature of a location candidate of the set of location candidates and a search feature associated with the location candidate of the set of location candidates.

13. The non-transitory computer readable medium of claim 12, wherein the location feature is independent of queries recorded in the query log.

14. The non-transitory computer readable medium of claim 13, wherein the location feature comprises at least one of a population of the location candidate, a population density of the location candidate, demography of the location candidate, a popularity of the location candidate, or a business population of the location candidate.

15. The non-transitory computer readable medium of claim 12, wherein the search feature is dependent on queries recorded in the query log and independent of searchers of the queries recorded in the query log.

16. The non-transitory computer readable medium of claim 9, wherein applying the machine learning technique to the training dataset to generate the location disambiguation model comprises generating a first location model for queries received from mobile devices and generating a second location model for queries received from fixed-location devices.

17. A system comprising:

a processor; and

a memory that stores instructions that, when executed by the processor, cause the processor to perform operations comprising

filtering a query log to identify a first query, the first query specifying an unambiguous location,

generating, from the first query, a second query having location ambiguity, wherein generating, from the first query, the second query having location ambiguity comprises

separating the first query into at least a first component and a second component, and

removing one of the first component or the second component, identifying a set of location candidates for the second query,

generating a training dataset based on comparing the unambiguous location specified in the first query with each location candidate of the set of location candidates for the second query,

applying a machine learning technique to the training dataset to generate a location disambiguation model, and

resolving location ambiguity in a query using the location disambiguation model.

18. The system of claim 17, wherein the training dataset further comprises a location feature of a location candidate of the set of location candidates and a search feature associated with the location candidate of the set of location candidates.

19. The system of claim 18, wherein the location feature comprises at least one of a population of the location candidate, a population density of the location candidate, demography of the location candidate, a popularity of the location candidate, or a business population of the location candidate.

20. The system of claim 17, wherein applying the machine learning technique to the training dataset to generate the location disambiguation model comprises generating a first location model for queries received from mobile devices and generating a second location model for queries received from fixed-location devices.