System and method for generating recommendations based on similarities between location information of multiple users
Various embodiments of a system and method for generating recommendations based on similarities between location information of multiple users are described. Various embodiments may include a location-based recommendation system configured to, for each given user of a group of users, determine the given user has traveled to one or more respective locations and determine one or more characteristics of the given user. The system may also determine that a particular user has traveled to or will travel to each of one or more particular locations. The system may further determine a similarity between the one or more particular locations and one or more locations to which specific ones of the group of users have traveled. The system may generate a recommendation for the particular user based on at least some of the determined characteristics of the specific ones of the group of users for which the similarity was determined.
1. A computer-implemented method, comprising: performing, by one or more hardware processors of one or more computers:
for individual given users of a plurality of users: receiving, by a location-based recommendation system, user location information that indicates a user's location, the location on a path associated with the user, and associated time information that the user was at the location for the given user from one or more computing devices associated with the given user, and determining one or more user characteristics for the given user;
determining, based on clustering distance metrics of multiple users of the plurality of users, one or more regions that define respective subsets of the plurality of users;
receiving, by the location-based recommendation system, user location information for a particular user from one or more computing devices associated with the particular user that indicates the user location in near real-time with the user's presence on the path;
determining, by a similarity analysis component, a particular subset of the subsets of the plurality of users that is similar to the particular user, wherein the determining the particular subset comprises determining a similarity between user path information comprising the user location information for the particular user and user path information of the plurality of users of the particular subset, wherein the similarity is based at least in part on a similarity between the real-time information for the particular user and time information indicating when one or more of the plurality of users of the particular subset were on a similar path associated with the particular subset;
for said particular user, generating, by a recommendation generation component, a recommendation based at least in part on the user characteristics corresponding to the particular subset of the plurality of users;
generating a message including the recommendation; and
transmitting said message over a network to at least one computing device controlled by the particular user such that the computing device receives the message in near real-time with the particular user's presence on the path, the message including the recommendation that is based at least in part on the user characteristics corresponding to the particular subset of the plurality of users.
2. The computer-implemented method of claim 1 , wherein the method further comprises determining one or more user characteristics for the particular user; wherein determining the particular subset of users further comprises determining a similarity between the user characteristics of the particular user and the user characteristics corresponding to the particular subset of the plurality of users.
3. The computer-implemented method of claim 1 , wherein said at least one computing device comprises one or more of: a mobile telephone, a personal digital assistant, a laptop computer system, a desktop computer system, and a navigation device.
4. The computer-implemented method of claim 1 , wherein the user location information for a given user specifies one or more of: a location to which that user has traveled, a location in which that user is currently traveling, a location to which that user is expected to travel, and a location to which that user is predicted to travel.
5. The computer-implemented method of claim 1 , wherein at least some of the user location information for the plurality of users and the particular user includes information generated based on one or more of: the global positioning system (GPS), communication tower triangulation, and user-input specifying a location.
6. The computer-implemented method of claim 1 , wherein said one or more user characteristics of the given user indicate one or more of: an item or service purchased by the given user, and a network-based activity performed by the given user.
7. The computer-implemented method of claim 6 , wherein said network-based activity includes one or more of: accessing an item detail page of an e-commerce web site, purchasing an item on an e-commerce web site, and adding an item to a wish list of items on an e-commerce website.
8. The computer-implemented method of claim 1 , wherein said one or more characteristics of the given user comprise an indication that the given user has traveled to a particular place of commerce.
9. The computer-implemented method of claim 1 , wherein said recommendation includes one or more of: a recommended item or service purchased by the particular subset of the plurality of users according to the user characteristics for the particular subset, and a recommended location determined from the user location information for the particular subset.
10. The computer-implemented method of claim 1 , wherein said determining the particular subset of the plurality of users that is similar to the particular user comprises weighting at least the user location information of the plurality of users with a time decay value that diminishes over a configurable time.
11. A computer-implemented method, comprising:
performing, by one or more hardware processors of one or more computers:
for individual given locations of a plurality of locations on one or more paths: determining, by a location-based recommendation system and for a plurality of users that have or are expected to visit respective ones of the individual given locations, behavior information including user location information, and associated time information that each of the plurality of users have or are expected to visit respective ones of the individual given locations, wherein at least some of said behavior information is received from one or more computing devices associated with the plurality of users;
for a particular user, receiving, by the location-based recommendation system, user location information indicating a particular location of said plurality of locations from a computing device associated with that user, wherein the user location information indicates the user location in near real-time with the user's presence on a path comprising the indicated particular location;
generating, by a recommendation generation component, a recommendation for the particular user, said recommendation specifying a recommended action based on a statistical analysis of the behavior information for the particular path, wherein the statistical analysis includes analysis of a similarity between the real-time information for the particular user and time information for one or more of the plurality of users;
wherein the recommendation comprises a recommendation of an item to be purchased through an e-commerce website;
generating a message comprising the recommendation; and
sending said message to at least one computing device controlled by the particular user such that the at least one computing device receives the message in near real-time with the particular user's presence on the path.
12. The computer-implemented method of claim 11 , wherein the recommended action of said recommendation indicates one or more of: an item to purchase, a service to purchase and a location to visit.
13. The computer-implemented method of claim 11 , wherein the user location information for the particular user specifies one or more of: a location to which that user has traveled, a location in which that user is currently traveling, a location to which that user is expected to travel, and a location to which that user is predicted to travel.
14. The computer-implemented method of claim 11 , wherein for a given one of the plurality of users, the behavior information of that user indicates one or more of: a location to which that user has traveled, a location in which that user is currently traveling, a location to which that user is expected to travel, and a location to which that user is predicted to travel, an item or service purchased by that user, and a network-based activity performed by that user.
15. The computer-implemented method of claim 11 , wherein said statistical analysis of the behavior information for the particular location comprises determining the recommended action as being a most popular action performed by the plurality of users in association with that particular location.
16. A system, comprising:
a memory; and
one or more hardware processors coupled to the memory, wherein the memory comprises program instructions that are executed by the one or more hardware processors to implement:
a location-based recommendation service configured to:
for individual given users of a plurality of users: receive user location information and associated time information that the user was at the location for the given user from one or more computing devices associated with the given user, the user location on a path associated with the user, and determine one or more user characteristics for the given user;
cluster distance metrics of multiple users of the plurality of users to determine one or more regions that define respective subsets of the plurality of users;
receive user location information for a particular user from one or more computing devices associated with the particular user, wherein the user location information for the particular user indicates the user location in near real-time with the user's presence on the path associated with the user;
determine a particular subset of the subsets of the plurality of users that are similar to the particular user, wherein the program instructions to determine the particular subset further comprise program instructions to determine a similarity between user path information for the particular user and user path information of the plurality of users of the particular subset, wherein the similarity is based at least in part on a similarity between the near real-time information for the particular user and the time information indicating when one or more of the plurality of users of the particular subset were on a similar path associated with the particular subset; and
for said particular user, generate a recommendation based on the user characteristics corresponding to the particular subset of the plurality of users;
wherein the program instructions are configured to:
generate a message comprising the recommendation, and
send said message to at least one computing device controlled by the particular user such that the computing device receives the message in near real-time with the particular user's presence on the path associated with the user, the message including the recommendation that is based at least in part on the user characteristics corresponding to the particular subset of the plurality of users.
17. The system of claim 16 , wherein the program instructions are configured to determine one or more user characteristics for the particular user; wherein to determine the particular subset of users, the program instructions are further configured to determine a similarity between the user characteristics of the particular user and the user characteristics corresponding to the particular subset of the plurality of users.
18. The system of claim 16 , wherein said at least one computing device comprises one or more of: a mobile telephone, a personal digital assistant, a laptop computer system, a desktop computer system, and a navigation device.
19. The system of claim 16 , wherein the user location information for a given user specifies one or more of: a location to which that user has traveled, a location in which that user is currently traveling, a location to which that user is expected to travel, and a location to which that user is predicted to travel.
20. The system of claim 16 , wherein at least some of the user location information for the plurality of users and the particular user includes information generated based on one or more of: the global positioning system (GPS), communication tower triangulation, and user-input specifying a location.
21. The system of claim 16 , wherein said one or more user characteristics of the given user indicate one or more of: an item or service purchased by the given user, and a network-based activity performed by the given user.
22. The system of claim 21 , wherein said network-based activity includes one or more of: accessing an item detail page of an e-commerce website, purchasing an item on an e-commerce web site, and adding an item to a wish list of items on an e-commerce web site.
23. The system of claim 16 , wherein said one or more characteristics of the given user comprise an indication that the given user has traveled to a particular place of commerce.
24. The system of claim 16 , wherein said recommendation includes one or more of: a recommended item or service purchased by the particular subset of the plurality of users according to the user characteristics for the particular subset, and a recommended location determined from the user location information for the particular sub set.
25. The system as recited in claim 16 , wherein the program instructions to determine the particular subset of the plurality of users that is similar to the particular user comprise program instructions to weight the user location information of the plurality of users with a time decay value that diminishes over a configurable time.
26. A non-transitory computer-readable storage medium, storing program instructions that are computer-executable to:
implement a location-based recommendation service configured to:
for individual given users of a plurality of users: receive user location information and associated time information that the user was at the location for the given user from one or more computing devices associated with the given user, the location on a path associated with the user, and determine one or more user characteristics for the given user;
cluster distance metrics of multiple users of the plurality of users to determine one or more regions that define respective subsets of the plurality of users;
determine, based at least in part on received user location information that indicates the user location in near real-time with the user's presence on the path, user path information for a particular user from one or more computing devices associated with the particular user;
determine a particular subset of the subsets of the plurality of users that is similar to the particular user, wherein the program instructions computer executable to determine the particular subset further comprise program instructions computer executable to determine a similarity between user path information of the particular user and user path information of the plurality of users of the particular subset, wherein the similarity is based at least in part on a similarity between the real-time information for the particular user and time information indicating when one or more of the plurality of users of the particular subset were on a similar path associated with the particular subset; and
for said particular user, generate a recommendation based at least in part on the user characteristics associated with behavior information for the location corresponding to the particular subset of the plurality of users;
wherein the program instructions when executed:
generate a message comprising the recommendation, and
send said message to at least one computing device controlled by the particular user such that the computing device receives the message in near real-time with the particular user's presence on the path associated with the user, the message including the recommendation that is based at least in part on the user characteristics corresponding to the particular subset of the plurality of users.
27. The non-transitory computer-readable storage medium of claim 26 , wherein the program instructions are configured to determine one or more user characteristics for the particular user; wherein to determine the particular subset of users, the program instructions are further configured to determine a similarity between the user characteristics of the particular user and the user characteristics corresponding to the particular subset of the plurality of users.
28. The non-transitory computer-readable storage medium of claim 26 , wherein said at least one computing device comprises one or more of: a mobile telephone, a personal digital assistant, a laptop computer system, a desktop computer system, and a navigation device.
29. The non-transitory computer-readable storage medium of claim 26 , wherein the user location information for a given user specifies one or more of: a location to which that user has traveled, a location in which that user is currently traveling, a location to which that user is expected to travel, and a location to which that user is predicted to travel.
30. The non-transitory computer-readable storage medium of claim 26 , wherein at least some of the user location information for the plurality of users and the particular user includes information generated based on one or more of: the global positioning system (GPS), communication tower triangulation, and user-input specifying a location.
31. The non-transitory computer-readable storage medium of claim 26 , wherein said one or more user characteristics of the given user indicate one or more of: an item or service purchased by the given user, and a network-based activity performed by the given user.
32. The non-transitory computer-readable storage medium of claim 31 , wherein said network-based activity includes one or more of: accessing an item detail page of an e-commerce web site, purchasing an item on an e-commerce web site, and adding an item to a wish list of items on an e-commerce web site.
33. The non-transitory computer-readable storage medium of claim 26 , wherein said one or more characteristics of the given user comprise an indication that the given user has traveled to a particular place of commerce.
34. The non-transitory computer-readable storage medium of claim 26 , wherein
said recommendation includes one or more of: a recommended item or service purchased by the particular subset of the plurality of users according to the user characteristics for the particular subset, and a recommended location determined from the user location information for the particular subset,
the recommendation is based at least in part upon a season in which the particular user will visit the location, and
the recommendation is based at least in part upon a season in which the particular subset of the plurality of users visited the location.
35. The non-transitory computer-readable storage medium of claim 26 , wherein the program instructions computer executable to determine the particular subset further comprise program instructions computer executable to weight the user location information of the plurality of users with a time decay value that diminishes over a configurable time.