Systems and methods for route prediction

Patent metadata

Field	Value
Publication number	US-11099024-B2
Application number	US-201815877651-A
Country	US
Kind code	B2
Filing date	Jan 23, 2018
Priority date	Nov 9, 2012
Publication date	Aug 24, 2021
Grant date	Aug 24, 2021

Abstract

Predicting the future location of a user based on predicting the route that the user might take is disclosed. The routes used by the user in the past are indexed to generate a dictionary of routes which can be further augmented with contextual data. The prior routes are encoded within the dictionary such that each term representing a respective one of the prior routes comprises a collection of unique identifiers wherein each of the unique identifiers represents a segment of the respective one of the prior routes. Techniques of text prediction, term frequency for dictionary scores and other language processing techniques are used to predict the further route of the user.

First claim

What is claimed is:

1. A method for route prediction using user mobile devices, comprising:
monitoring, with at least one processor, locations of a plurality of mobile devices associated with a plurality of users based on location data of each mobile device;
generating, with at least one processor, a plurality of routes for a user of the plurality of users based on the locations of a mobile device operated by the user, each route of the plurality of routes comprising a plurality of segments;
determining, with at least one processor, route frequency data for at least a portion of routes of the plurality of routes, the route frequency data comprising a count of traversals of each route with the mobile device;
determining, with at least one processor, route context data for at least a portion of routes of the plurality of routes, the route context data comprising contextual data relating to each traversal of each route with the mobile device;
generating, with at least one processor, a route database based on the plurality of routes, the route frequency data, and the route context data by:
determining routes frequently traversed by the user from the plurality of routes based on the route frequency data;
selecting a set of top routes from the routes frequently traversed by the user based on locations of commercial establishments and residential areas with respect to each route of the routes frequently traversed, such that routes that predominantly pass through a residential area are not selected; and
encoding the set of top routes with a plurality of identifiers, wherein each route of the set of top routes is encoded as a sequence of identifiers of the plurality of identifiers, each identifier uniquely identifying a segment of the route;

detecting, with at least one processor, that the mobile device has partially traversed a subset of routes of the set of top routes based on a comparison between a partially traversed route and the set of top routes, such that the partially traversed route matches a first portion of the sequence of identifiers for each route of the subset of routes; and
determining, with at least one processor, a predicted route from the subset of routes based at least partially on at least one of the route frequency data and the route context data.

2. The method of claim 1 , wherein determining the predicted route from the subset of routes comprises selecting the predicted route based on the route frequency data, such that the predicted route comprises a greater count of traversals compared to other routes of the subset of routes.

3. The method of claim 1 , wherein determining the predicted route from the subset of routes comprises:
determining, with at least one processor, a likelihood that the user will traverse each route of the subset of routes based at least partially on the route context data; and
selecting the predicted route based on a greater likelihood that the user will traverse the predicted route than other routes of the subset of routes.

4. The method of claim 3 , wherein the route context data comprises a time of day, and wherein the likelihood that the user will traverse each route of the subset of routes is based at least partially on comparing a current time of day to the route context data for each of the routes of the subset of routes.

5. The method of claim 3 , wherein determining the likelihood that the user will traverse each route of the subset of routes is based at least partially on the route context data and the route frequency data.

6. The method of claim 1 , further comprising:
determining, with at least one processor, at least one offer that is available at a merchant located along the predicted route; and
communicating the at least one offer to the mobile device while the mobile device traverses the predicted route.

7. The method of claim 1 , further comprising automatically configuring a starting location of the user by analyzing, with at least one processor, the locations of the mobile device.

8. The method of claim 7 , wherein detecting that the mobile device has partially traversed a subset of routes of the set of top routes is based on the starting location of the user.

9. A system for route prediction using user mobile devices, comprising at least one server computer including at least one processor, the at least one server computer programmed or configured to:
monitor locations of a plurality of mobile devices associated with a plurality of users based on location data of each mobile device;
generate a plurality of routes for a user of the plurality of users based on the locations of a mobile device operated by the user, each route of the plurality of routes comprising a plurality of segments;
determine route frequency data for at least a portion of routes of the plurality of routes, the route frequency data comprising a count of traversals of each route with the mobile device;
determine route context data for at least a portion of routes of the plurality of routes, the route context data comprising contextual data relating to each traversal of each route with the mobile device;
generate a route database based on the plurality of routes, the route frequency data, and the route context data by:
determining routes frequently traversed by the user from the plurality of routes based on the route frequency data;
selecting a set of top routes from the routes frequently traversed by the user based on locations of commercial establishments and residential areas with respect to each route of the routes frequently traversed, such that routes that predominantly pass through a residential area are not selected; and
encoding the set of top routes with a plurality of identifiers, wherein each route of the set of top routes is encoded as a sequence of identifiers of the plurality of identifiers, each identifier uniquely identifying a segment of the route;

detect that the mobile device has partially traversed a subset of routes of the set of top routes based on a comparison between a partially traversed route and the set of top routes, such that the partially traversed route matches a first portion of the sequence of identifiers for each route of the subset of routes; and
determine a predicted route from the subset of routes based at least partially on at least one of the route frequency data and the route context data.

10. The system of claim 9 , wherein the at least one server computer is configured to determine the predicted route from the subset of routes by selecting the predicted route based on the route frequency data, such that the predicted route comprises a greater count of traversals compared to other routes of the subset of routes.

11. The system of claim 9 , wherein the at least one server computer is further configured to:
determine a likelihood that the user will traverse each route of the subset of routes based at least partially on the route context data; and
select the predicted route based on a greater likelihood that the user will traverse the predicted route than other routes of the subset of routes.

12. The system of claim 11 , wherein the route context data comprises a time of day, and wherein the likelihood that the user will traverse each route of the subset of routes is based at least partially on comparing a current time of day to the route context data for each of the routes of the subset of routes.

13. The system of claim 11 , wherein determining the likelihood that the user will traverse each route of the subset of routes is based at least partially on the route context data and the route frequency data.

14. The system of claim 11 , wherein the at

Assignees

• Visa Int Service Ass

Inventors

• Spears Joseph
• Carlson Mark
• Stan Patrick

Classifications

• G01C21/3484 Primary

  Personalized, e.g. from learned user behaviour or user-defined profiles · CPC title