Location awareness using local semantic scoring

• US 9,471,693 B2
• Filed: 05/29/2013
• Issued: 10/18/2016
• Est. Priority Date: 05/29/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Mishra, Pragyana K., Reddy, Nagareddy S.
• Primary Examiner(s): Lu, Kuen
• Application Number:
  US13/905,025
• Publication Number:
  US 20140358897A1
• Time in Patent Office: 1,238 Days
• Field of Search: 707/722
• US Class Current: 1/1
• 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

Abstract

Systems and methods are provided for determining continuous scoring models that describe one or more features regarding a geographic area. The scoring models can be fit to incorporate information from both point sources as well as area information. Based on the scoring models, a user can obtain feature values for the one or more features at any desired location or for a desired geographic area. Because the features are represented by a continuous function, comparisons of feature values for arbitrary geographic areas and/or combinations of features to form secondary values can also be performed.

Claims

1. A method for presenting feature scores related to a geographic area, comprising:

determining a feature score location;

receiving a selection of one or more features, each selected feature being associated with at least one scoring model, a scoring model associated with a selected feature comprising a smooth, continuous scoring model defined over a plurality of basis grid areas, the determined feature score location being contained within the plurality of basis grid areas, the scoring model being continuous at a boundary between basis grid areas;

sampling the at least one scoring model associated with each selected feature to determine feature scores for the one or more selected features;

displaying a map that includes at least a portion of the determined feature score location; and

presenting the feature scores of the selected features.

2. The method of claim 1, wherein the feature score location comprises at least one of a point or an area within a bounding shape.

3. The method of claim 1, wherein defining a continuous scoring model over a basis grid area comprises fitting the continuous scoring model to one or more features within the basis grid area, fitting the continuous scoring model to one or more basis functions associated with features in a basis grid area, or a combination thereof.

4. The method of claim 3, wherein defining the continuous scoring model over a basis grid area further comprises fitting the continuous scoring model in a first basis grid area to one or more basis functions associated with features in a second basis grid area, the second basis grid area being different from the first basis grid area.

5. The method of claim 1, wherein one or more features within a basis grid area comprise social media features, at least one social media feature corresponding to a check-in feature for an entity or geographic area, a like feature for an entity or geographic area, or a combination thereof.

6. The method of claim 1, wherein sampling the scoring model comprises integrating the scoring model over an area corresponding to the determined feature score location, and dividing the integrated value by the area corresponding to the determined feature score location, or wherein sampling the scoring model comprises determining a value of the scoring model at a plurality of points within a bounding shape corresponding to the feature score location.

7. The method of claim 6, wherein presenting the feature scores of a selected feature comprises presenting the values of the scoring model at two or more of the plurality of points within the bounding shape.

8. The method of claim 6, wherein sampling the scoring model further comprises summing the scoring model values at the plurality of points and dividing by a number corresponding to the plurality of points.

9. The method of claim 1, wherein determining a feature score location comprises selecting a location associated with a responsive result to a search query.

10. The method of claim 1, wherein presenting a feature score of a selected feature comprises displaying the feature score and an indicator of a range of possible feature scores.

11. The method of claim 1, wherein at least one selected feature comprises a secondary value, the feature score for the secondary value being determined based on sampling of two or more scoring models.

12. The method of claim 1, wherein the continuous scoring model defined over the plurality of basis grid areas comprises a smooth, continuous scoring model defined over the plurality of basis grid areas.

13. A system for presenting feature scores related to a geographic area, comprising:

a processor and associated memory, the processor executing computer-executable instructions to;

determine a sampling area;

receive a selection of one or more features, each selected feature being associated with at least one scoring model, a scoring model comprising a smooth, continuous scoring model defined over a plurality of basis grid areas, the determined sampling area being contained within the plurality of basis grid areas;

receive feature thresholds for the one or more selected features;

sample the at least one scoring model associated with each selected feature to identify at least one feature score location within the sampling area having feature scores for the one or more selected features that satisfy the received feature thresholds;

display a map that includes the at least one feature score location; and

present an indicator of the at least one identified feature score location on the displayed map.

14. The system of claim 13, wherein the sampling area comprises a pre-defined geographic area, an area within a bounding shape, or a combination thereof.

15. The system of claim 13, wherein a received feature threshold is satisfied based on a feature score being greater than a first threshold value, a feature score being less than a second threshold value, or a combination thereof.

16. The system of claim 13, wherein presenting an indicator of the at least one feature score location comprises displaying areas corresponding to areas where each feature satisfies a feature threshold.

17. A method for presenting feature scores related to a geographic area, comprising:

acquiring data for one or more features within a plurality of basis grid areas;

generating, for each feature of the one or more features, a scoring model based on the acquired feature data in the plurality of basis grid areas, the scoring model comprising a smooth, continuous scoring model within the plurality of basis grid areas, at least one basis grid area comprising multiple points of acquired feature data, at least one Gaussian function being used to represent the acquired feature data;

determining a feature score location, the determined feature score location being contained within the one or more basis grid areas;

receiving a selection of a feature;

sampling the scoring model associated with the selected feature to determine a feature score;

displaying a map that includes at least a portion of the determined feature score location; and

presenting a feature score based on the determined feature score.

18. The method of claim 17, wherein the scoring model is generated based on fitting the scoring model to one or more basis functions associated with features in a basis grid area.

19. The method of claim 17, wherein acquiring data within a plurality of basis grid areas comprises acquiring one-dimensional or two-dimensional data.

20. The method of claim 17, wherein the scoring model is continuous at a boundary between basis grid areas.