Place Affinity Estimation

• US 20130151148A1
• Filed: 03/30/2012
• Published: 06/13/2013
• Est. Priority Date: 12/12/2011
• Status: Active Grant

• Current Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Original Assignee: Toyota Infotechnology Center Co., Ltd. (Toyota Motor Corporation), Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Inventors: Oguchi, Kentaro, Parundekar, Rahul Ravi, Takeuchi, Shojiro
• Granted Patent: US 9,026,364 B2
• Time in Patent Office: Days
• US Class Current: 701/521
• CPC Class Codes: G01C 21/3679 Retrieval, searching and output of POI information, e.g. hotels, restaurants, shops, filling stations, parking facilities G01C21/3611 takes precedence

Abstract

A system and method for estimating a place affinity for a user is disclosed. The system comprises a gathering module, a communication module and a scoring module. The gathering module receives a place and retrieves rich place data associated with the place. The communication module retrieves user profile data associated with a user and a place affinity model associated with the user. The scoring module estimates an affinity score for the place using the place affinity model based at least in part on the rich place data and the user profile data.

Claims

1. A computer-implemented method, comprising:

receiving sensor data for a vehicle associated with a user, the sensor data describing a current vehicle location;

retrieving candidate place data describing one or more candidate places within a range of the vehicle location;

retrieving historical place data describing one or more places visited by the user in the past; and

estimating a currently visited place for the user based at least in part on the one or more of the current vehicle location, the candidate place data and the historical place data.

2. The method of claim 1 further comprising:

estimating a first time period spent at the estimated currently visited place;

retrieving a visiting frequency for the estimated currently visited place and a second time period spent on a past visit to the estimated currently visited place; and

determining like data for the estimated currently visited place, the like data indicating if the user likes the estimated currently visited place.

3. The method of claim 1, wherein estimating a currently visited place for the user based at least in part on the one or more of the current vehicle location, the candidate place data and the historical place data comprises choosing a place from the one or more candidate places as the currently visited place based on a current time of the day and a time period spent at the currently visited place.

4. The method of claim 1, wherein estimating a currently visited place for the user based at least in part on the one or more of the current vehicle location, the candidate place data and the historical place data comprises selecting among the one or more candidate places a place within a walkable distance to the current vehicle location as the currently visited place.

5. The method of claim 1, wherein the currently visited place is one of the one or more candidate places and also one of the one or more places visited by the user in the past.

6. A system, comprising:

a communication module for receiving sensor data for a vehicle associated with a user, the sensor data describing a current vehicle location, the communication module retrieving candidate place data describing one or more candidate places within a range of the current vehicle location, the communication module retrieving historical place data describing one or more places visited by the user in the past; and

a place estimation module communicatively coupled to the communication module for receiving one or more of the sensor data, the candidate place data and the historical place data, the place estimation module estimating a currently visited place for the user based at least in part on the one or more of the current vehicle location, the candidate place data and the historical place data.

7. The system of claim 6 further comprising a sensor communicatively coupled to the communication module, the sensor estimating a first time period spent at the estimated currently visited place.

8. The system of claim 7 further comprising a liking module communicatively coupled to the communication module and the sensor, the liking module receiving the first time period spent at the estimated currently visited place and retrieving a visiting frequency for the estimated currently visited place and a second time period spent on a past visit to the estimated currently visited place, the liking module further determining like data for the estimated currently visited place, the like data indicating if the user likes the estimated currently visited place.

9. The system of claim 6, wherein estimating a currently visited place for the user based at least in part on the one or more of the current vehicle location, the candidate place data and the historical place data comprises choosing a place from the one or more candidate places as the currently visited place based on a current time of the day and a time period spent at the currently visited place.

10. The system of claim 6, wherein estimating a currently visited place for the user based at least in part on the one or more of the current vehicle location, the candidate place data and the historical place data comprises selecting among the one or more candidate places a place within a walkable distance to the current vehicle location as the currently visited place.

11. The system of claim 6, wherein the currently visited place is one of the one or more candidate places and also one of the one or more places visited by the user in the past.

12. A computer program product comprising a non-transitory computer readable medium encoding instructions that, in response to execution by a computing device, cause the computing device to perform operations comprising:

receiving sensor data for a vehicle associated with a user, the sensor data describing a current vehicle location;

retrieving candidate place data describing one or more candidate places within a range of the current vehicle location;

retrieving historical place data describing one or more places visited by the user in the past; and

estimating a currently visited place for the user based at least in part on the one or more of the current vehicle location, the candidate place data and the historical place data.

13. The computer program product of claim 12, wherein instructions encoded in the computer readable medium when executed cause the computing device to perform operations further comprising:

estimating a first time period spent at the estimated currently visited place;

retrieving a visiting frequency for the estimated currently visited place and a second time period spent on a past visit to the estimated currently visited place; and

determining like data for the estimated currently visited place, the like data indicating if the user likes the estimated currently visited place.

14. The computer program product of claim 12, wherein estimating a currently visited place for the user based at least in part on the one or more of the current vehicle location, the candidate place data and the historical place data comprises choosing a place from the one or more candidate places as the currently visited place based on a current time of the day and a time period spent at the currently visited place.

15. The computer program product of claim 12, wherein estimating a currently visited place for the user based at least in part on the one or more of the current vehicle location, the candidate place data and the historical place data comprises selecting among the one or more candidate places a place within a walkable distance to the current vehicle location as the currently visited place.

16. The computer program product of claim 12, wherein the currently visited place is one of the one or more candidate places and also one of the one or more places visited by the user in the past.

44. The method of claim 14, wherein personalizing the one or more matching places based at least in part on the one or more affinity scores further comprises:

ordering the one or more matching places based at least in part on the one or more affinity scores; and

selecting one or more matching places with highest affinity scores.

17. A computer-implemented method, comprising:

receiving a place associated with a user;

retrieving rich place data associated with the place;

retrieving user profile data for the user;

retrieving a previous state of a place affinity model associated with the user;

generating the place affinity model associated with the user based at least in part on the rich place data and the previous state of the place affinity model.

18. The method of claim 17, wherein the place is marked with an affinity value by the user based at least in part on inputs from one or more of an interactive dialog, a survey and retrieved personal data from one or more social networks.

19. The method of claim 17, wherein the rich place data comprises one or more of historical place data for the place, social data describing one or more social connections of the user that have visited the place and web data describing one or more reviews for the place.

20. The method of claim 17, wherein the place affinity model comprises model data describing one or more favorite places for the user and one or more common characteristics for the one or more favorite places.

21. The method of claim 17 further comprising:

receiving sensor data describing a current time of the day and a current vehicle location; and

determining a situation for the user based at least in part on the sensor data,

wherein the situation for the user and the sensor data are used to estimate the place affinity model.

22. The method of claim 17 further comprising:

receiving a new place;

retrieving new rich place data associated with the new place;

retrieving the place affinity model associated with the user; and

updating the place affinity model associated with the user based at least in part on the new rich place data.

23. A system, comprising:

a communication module for receiving a place associated with a user, the communication module retrieving user profile data for the user;

a gathering module communicatively coupled to the communication module for retrieving rich place data associated with the place; and

a modeling module communicatively coupled to the gathering module for receiving the rich place data, the modeling module retrieving a previous state of a place affinity model associated with the user and generating the place affinity model associated with the user based at least in part on the rich place data and the previous state of the place affinity model.

24. The system of claim 23, wherein the place is marked with an affinity value by the user based at least in part on inputs from one or more of an interactive dialog, a survey and retrieved personal data from one or more social networks.

25. The system of claim 23, wherein the rich place data comprises one or more of historical place data for the place, social data describing one or more social connections of the user that have visited the place and web data describing one or more reviews for the place.

26. The system of claim 23, wherein the place affinity model comprises model data describing one or more favorite places for the user and one or more common characteristics for the one or more favorite places.

27. The system of claim 23, wherein the communication module receives sensor data describing a current time of the day and a current vehicle location and the modeling module determines a situation for the user based at least in part on the sensor data, wherein the situation for the user and the sensor data are used to estimate the place affinity model.

28. The system of claim 23, wherein the gathering module receives a new place and retrieves new rich place data associated with the new place.

29. The system of claim 28 further comprising a updating module communicatively coupled to the gathering module, the updating module retrieving the place affinity model associated with the user and updating the place affinity model associated with the user based at least in part on the new rich place data.

30. A computer program product comprising a non-transitory computer readable medium encoding instructions that, in response to execution by a computing device, cause the computing device to perform operations comprising:

receiving a place associated with a user;

retrieving rich place data associated with the place;

retrieving user profile data for the user;

retrieving a previous state of a place affinity model associated with the user; and

generating the place affinity model associated with the user based at least in part on the rich place data and the previous state of the place affinity model.

31. The computer program product of claim 30, wherein the place is marked with an affinity value by the user based at least in part on inputs from one or more of an interactive dialog, a survey and retrieved personal data from one or more social networks.

32. The computer program product of claim 30, wherein the rich place data comprises one or more of historical place data for the place, social data describing one or more social connections of the user that have visited the place and web data describing one or more reviews for the place.

33. The computer program product of claim 30, wherein the place affinity model comprises model data describing one or more favorite places for the user and one or more common characteristics for the one or more favorite places.

34. The computer program product of claim 30, wherein instructions encoded in the computer readable medium when executed cause the computing device to perform operations further comprising:

receiving sensor data describing a current time of the day and a current vehicle location; and

determining a situation for the user based at least in part on the sensor data,

wherein the situation for the user and the sensor data are used to estimate the place affinity model.

35. The computer program product of claim 30, wherein instructions encoded in the computer readable medium when executed cause the computing device to perform operations further comprising:

receiving a new place;

retrieving new rich place data associated with the new place;

retrieving the place affinity model associated with the user; and

updating the place affinity model associated with the user based at least in part on the new rich place data.

36. A computer-implemented method, comprising:

receiving a place;

retrieving user profile data associated with a user and a place affinity model associated with the user;

retrieving rich place data associated with the place; and

estimating an affinity score for the place using the place affinity model based at least in part on the rich place data and the user profile data.

37. The method of claim 36 further comprising:

receiving sensor data describing a current time of the day and a current vehicle location; and

determining a situation for the user based at least in part on the sensor data,

wherein the situation for the user and the sensor data are used to generate the affinity score for the place.

38. A system, comprising:

a gathering module for receiving a place, the gathering module further retrieving rich place data associated with the place;

a communication module communicatively coupled to the gathering module, the communication module retrieving user profile data associated with a user and a place affinity model associated with the user; and

a scoring module communicatively coupled to the communication module and the gathering module for receiving one or more of the rich place data, the user profile data and the place affinity model associated with the user, the scoring module estimating an affinity score for the place using the place affinity model based at least in part on the rich place data and the user profile data.

39. The system of claim 38, wherein the communication module receives sensor data describing a current time of the day and a current vehicle location.

40. The system of claim 39 further comprising a modeling module communicatively coupled to the communication module, the modeling module determining a situation for the user based at least in part on the sensor data, wherein the situation for the user and the sensor data are used to generate the affinity score for the place.

41. A computer program product comprising a non-transitory computer readable medium encoding instructions that, in response to execution by a computing device, cause the computing device to perform operations comprising:

receiving a place;

retrieving user profile data associated with a user and a place affinity model associated with the user;

retrieving rich place data associated with the place; and

estimating an affinity score for the place using the place affinity model based at least in part on the rich place data and the user profile data.

42. The computer program product of claim 41, wherein instructions encoded in the computer readable medium when executed cause the computing device to perform operations further comprising:

receiving sensor data describing a current time of the day and a current vehicle location; and

determining a situation for the user based at least in part on the sensor data,

wherein the situation for the user and the sensor data are used to generate the affinity score for the place.

43. A computer-implemented method, comprising:

receiving a search request from a user;

determining one or more matching places matched to the search request;

estimating one or more affinity scores for the one or more matching places; and

personalizing the one or more matching places based at least in part on the one or more affinity scores.

45. A system, comprising:

a communication module for receiving a search request from a user;

a personalization module communicatively coupled to the communication module for determining one or more matching places matched to the search request, the personalization module personalizing the one or more matching places based at least in part on one or more affinity scores; and

a scoring module communicatively coupled to the personalization module for estimating the one or more affinity scores for the one or more matching places, the scoring module sending the one or more affinity scores for the one or more matching places to the personalization module.

46. The system of claim 45, wherein personalizing the one or more matching places based at least in part on the one or more affinity scores further comprises:

ordering the one or more matching places based at least in part on the one or more affinity scores; and

selecting one or more matching places with highest affinity scores.

47. A computer program product comprising a non-transitory computer readable medium encoding instructions that, in response to execution by a computing device, cause the computing device to perform operations comprising:

receiving a search request from a user;

determining one or more matching places matched to the search request;

estimating one or more affinity scores for the one or more matching places; and

personalizing the one or more matching places based at least in part on the one or more affinity scores.

48. The computer program product of claim 47, wherein personalizing the one or more matching places based at least in part on the one or more affinity scores further comprises:

ordering the one or more matching places based at least in part on the one or more affinity scores; and

selecting one or more matching places with highest affinity scores.

49. A computer-implemented method, comprising:

receiving a place associated with a user;

retrieving rich place data associated with the place;

retrieving user profile data for the user;

retrieving a default standard-user place affinity model pre-computed for one or more groups of users; and

generating a place affinity model associated with the user based at least in part on the rich place data and the default standard-user place affinity model.

50. A computer-implemented method, comprising:

receiving a place associated with a user;

retrieving rich place data associated with the place;

retrieving user profile data for the user;

retrieving an average-user place affinity model generated by aggregating individual place affinity models for a group of users; and

generating the place affinity model associated with the user based at least in part on the rich place data and the average-user place affinity model.