Method and system for recommending multi-modal itineraries

Patent metadata

Field	Value
Publication number	US-11544638-B2
Application number	US-201916408683-A
Country	US
Kind code	B2
Filing date	May 10, 2019
Priority date	May 10, 2019
Publication date	Jan 3, 2023
Grant date	Jan 3, 2023

Abstract

Itineraries may be recommended using rider context information relating to a rider. The rider context information may be obtained from various sources. Location information including an origin and a destination may obtained. A set of itineraries may be generated based on the rider context information. Each itinerary may include at least one mode of transportation to allow the rider to travel from the origin to the destination. Each itinerary in the set of itineraries may be ranked based on the rider context information.

First claim

What is claimed is:

1. A system, comprising one or more processors and one or more non-transitory computer-readable memories coupled to the one or more processors and configured with instructions executable by the one or more processors to cause the system to perform operations comprising:
obtaining historical rider requests for rides and historical itinerary selections, wherein the historical rider requests comprise historical rider context information and historical location information associated with each historical rider request, and the historical itinerary selections comprise a set of historical itineraries generated by a ride-hailing platform and a selection from the set of historical itineraries;
training a machine learning model based on the historical rider requests and the historical itinerary selections for generating scores for itinerary candidates in response to a given rider's request for a ride, wherein the scores represent likelihoods of the given rider selecting corresponding itinerary candidates;
obtaining rider context information associated with a rider's request for a ride;
obtaining location information, wherein the location information includes an origin and a destination;
generating a set of itineraries, wherein each itinerary in the set of itineraries includes at least one mode of transportation to allow a rider to travel from the origin to the destination;
feeding the rider context information and the set of itineraries into the trained machine learning model to obtain a plurality of scores;
ranking each itinerary in the set of itineraries based on the plurality of scores;
receiving a rider selection from the set of itineraries; and
retraining the machine learning model based on the received rider selection and the rider context information.

2. The system of claim 1 , wherein the rider context information includes at least one of: demographic information and environmental data.

3. The system of claim 1 , wherein the rider context information includes timing information indicating a time for a future trip time.

4. The system of claim 1 , wherein the rider context information includes filtering criteria selected by the rider, and wherein each itinerary in the generated set of itineraries satisfies the filtering criteria.

5. The system of claim 1 , wherein the at least one mode of transportation includes at least one of: a self-controlled transportation mode and a non-self-controlled transportation mode.

6. A computer-implemented method, comprising:
obtaining, by a computing device, a trained machine learning model for generating scores for itinerary candidates in response to a given rider's request for a ride, wherein the scores represent likelihoods of the given rider selecting corresponding itinerary candidates, wherein the trained machine learning model is trained based on historical rider requests for rides and historical itinerary selections, wherein the historical rider requests comprise historical rider context information and historical location information associated with each historical rider request, and the historical itinerary selections comprise a set of historical itineraries generated by a ride-hailing platform and a selection from the set of historical itineraries;
obtaining, by the computing device, rider context information associated with a rider's request for a ride;
obtaining, by the computing device, location information, wherein the location information includes an origin and a destination;
generating, by the computing device, a set of itineraries, wherein each itinerary in the set of itineraries includes at least one mode of transportation to allow a rider to travel from the origin to the destination;
feeding, by the computing device, the rider context information and the set of itineraries into the trained machine learning model to obtain a plurality of scores;
ranking, by the computing device, each itinerary in the set of itineraries based on plurality of scores;
receiving, by the computing device, a rider selection from the set of itineraries; and
retraining, by the computing device, the machine learning model based on the received rider selection and the rider context information.

7. The method of claim 6 , wherein the rider context information includes at least one of: demographic information and environmental data.

8. The method of claim 6 , wherein the rider context information includes filtering criteria selected by the rider, and wherein each itinerary in the generated set of itineraries satisfies the filtering criteria.

9. The method of claim 6 , wherein the at least one mode of transportation includes at least one of: a self-controlled transportation mode and a non-self-controlled transportation mode.

10. The method of claim 6 , wherein the rider context information includes timing information indicating a time for a future trip time.

11. The method of claim 6 , wherein the rider context information includes filtering criteria selected by the rider, and wherein each itinerary in the generated set of itineraries satisfies the filtering criteria.

12. The method of claim 6 , wherein the at least one mode of transportation includes at least one of: a self-controlled transportation mode and a non-self-controlled transportation mode.

13. A non-transitory computer-readable storage medium configured with instructions executable by one or more processors to cause the one or more processors to perform operations comprising:
obtaining a trained machine learning model for generating scores for itinerary candidates in response to a given rider's request for a ride, wherein the scores represent likelihoods of the given rider selecting corresponding itinerary candidates, wherein the trained machine learning model is trained based on historical rider requests for rides and historical itinerary selections, wherein the historical rider requests comprise historical rider context information and historical location information associated with each historical rider request, and the historical itinerary selections comprise a set of historical itineraries generated by a ride-hailing platform and a selection from the set of historical itineraries;
obtaining rider context information associated with a rider's request for a ride;
obtaining location information, wherein the location information includes an origin and a destination;
generating a set of itineraries, wherein each itinerary in the set of itineraries includes at least one mode of transportation to allow a rider to travel from the origin to the destination;
feeding the rider context information and the set of itineraries into the trained machine learning model to obtain a plurality of scores;
ranking each itinerary in the set of itineraries based on the plurality of scores;
receiving a rider selection from the set of itineraries; and
retraining the machine learning model based on the received rider selection and the rider context information.

14. The non-transitory computer-readable storage medium of claim 13 , wherein the rider context information includes at least one of: demographic information and environmental data.

15. The non-transitory computer-readable storage medium of claim 13 , wherein the rider context information includes timing information indicating a time for a future trip time.

16. The non-transitory computer-readable storage medium of claim 13 , wherein the rider context information includes filtering criteria selected by the rider, and wherein each itinerary in the generated set of itineraries satisfies the filtering criteria.

17. The non-tr

Assignees

• Beijing Didi Infinity Technology & Dev Co Ltd

Inventors

• Tan Bo
• Qi Shanxiang
• Pan Bin

Classifications

• G06Q10/02

  Reservations, e.g. for tickets, services or events · CPC title

• G06N20/00

  Machine learning · CPC title

• G06Q10/025 Primary

  Coordination of plural reservations, e.g. plural trip segments, transportation combined with accommodation · CPC title

• G06Q50/30

  Physics · mapped topic

• G06Q50/40

  Business processes related to the transportation industry (shipping G06Q10/083) · CPC title