Cognitive route quality-learning service

1. A computer-implemented route-quality learning method, the method comprising:

collecting a first route for a trip recommended from a first third-party application;

collecting a second route for the trip recommended from a second third-party application;

comparing the current location of a user with a destination of the user as the user travels along the first route or the second route during the trip;

detecting, via map-matching or geo-fencing, an arrival of the user at the destination during the trip;

responsive to detecting the arrival of the user at the destination during the trip, estimating sub-optimal alternatives for the trip given completion data of the trip using both the first third-party application and the second third-party application recommendation, the sub-optimal alternatives being based on a route starting at the same time as the first route and the second route, being based on an entirety of the trip being completed from a start point to an end point of the trip, and the sub-optimal alternatives being routes that have a longer trip time than the first route and the second route at the time of taking the trip; and

during a second trip by the user at a later time after the trip is completed, recommending one of the sub-optimal alternatives for the trip as a route to take for the second trip,

wherein an explanation of a set of features are provided which differentiate an optimal route from the sub-optimal alternatives, and wherein the first route, the second route, and the sub-optimal alternatives are displayed for the user in a format indicating a time saved by taking the one of the sub-optimal alternatives.

2. The method of claim 1 , wherein the alternatives are estimated given current conditions including:

a traffic condition;

an incident report;

an incentive offered on each route; and

a user preference.

3. The method of claim 1 , further comprising updating a calculation factor of the sub-optimal alternatives based on a user preference for a first type of alternative route over a second type of alternative route.

4. The method of claim 1 , further comprising updating the sub-optimal alternatives for the user according to user preferences.

5. The method of claim 1 , wherein the sub-optimal alternatives are determined using a belief system that is updated for given historical realizations once the trip is completed.

6. The method of claim 5 , wherein a CHAID tree is calibrated to provide a key feature of present conditions that alters the sub-optimal alternatives from previously provided alternative routes.

7. The method of claim 1 , further comprising providing a reason that a first alternative route is ranked higher than a second alternative route.

8. The computer-implemented method of claim 1 , performed in a cloud-computing environment.

9. A computer program product for route-quality learning, the computer program product comprising a computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform:

collecting a first route for a trip recommended from a first third-party application;

collecting a second route for the trip recommended from a second third-party application;

comparing the current location of a user with a destination of the user as the user travels along the first route or the second route during the trip;

detecting, via map-matching or geo-fencing, an arrival of the user at the destination during the trip;

responsive to detecting the arrival of the user at the destination during the trip, estimating sub-optimal alternatives for the trip given completion data of the trip using both the first third-party application and the second third-party application recommendation, the sub-optimal alternatives being based on a route starting at the same time as the first route and the second route, being based on an entirety of the trip being completed from a start point to an end point of the trip, and the sub-optimal alternatives being routes that have a longer trip time than the first route and the second route at the time of taking the trip; and

during a second trip by the user at a later time after the trip is completed, recommending one of the sub-optimal alternatives for the trip as a route to take for the second trip,

wherein an explanation of a set of features are provided which differentiate an optimal route from the sub-optimal alternatives, and wherein the first route, the second route, and the sub-optimal alternatives are displayed for the user in a format indicating a time saved by taking the one of the sub-optimal alternatives.

10. The computer program product of claim 9 , wherein the alternatives are estimated given current conditions including:

a traffic condition;

an incident report;

an incentive offered on each route; and

a user preference.

11. The computer program product of claim 9 , further comprising updating a calculation factor of the sub-optimal alternatives based on a user preference for a first type of alternative route over a second type of alternative route.

12. The computer program product of claim 9 , further comprising updating the sub-optimal alternatives for the user according to user preferences.

13. The computer program product of claim 9 , wherein the sub-optimal alternatives are determined using a belief system that is updated for given historical realizations once the trip is completed.

14. The computer program product of claim 13 , wherein a CHAID tree is calibrated to provide a key feature of present conditions that alters the sub-optimal alternatives from previously provided alternative routes.

15. The computer program product of claim 9 , further comprising providing a reason that a first alternative route is ranked higher than a second alternative route.

16. A route-quality learning system, the system comprising:

a processor; and

a memory, the memory storing instructions to cause the processor to perform:

collecting a first route for a trip recommended from a first third-party application;

collecting a second route for the trip recommended from a second third-party application;

comparing the current location of a user with a destination of the user as the user travels along the first route or the second route during the trip;

detecting, via map-matching or geo-fencing, an arrival of the user at the destination during the trip;

responsive to detecting the arrival of the user at the destination during the trip, estimating sub-optimal alternatives for the trip given completion data of the trip using both the first third-party application and the second third-party application recommendation, the sub-optimal alternatives being based on a route starting at the same time as the first route and the second route, being based on an entirety of the trip being completed from a start point to an end point of the trip, and the sub-optimal alternatives being routes that have a longer trip time than the first route and the second route at the time of taking the trip; and

during a second trip by the user at a later time after the trip is completed, recommending one of the sub-optimal alternatives for the trip as a route to take for the second trip,

wherein an explanation of a set of features are provided which differentiate an optimal route from the sub-optimal alternatives, and wherein the first route, the second route, and the sub-optimal alternatives are displayed for the user in a format indicating a time saved by taking the one of the sub-optimal alternatives.

17. The system of claim 16 , wherein the alternatives are estimated given current conditions including:

a traffic condition;

an incident report;

an incentive offered on each route; and

a user preference.

18. The system of claim 16 , embodied in a cloud-computing environment.