Systems and methods for bot selection calibration in two-way communications
The present disclosure relates generally to facilitating two-way communications. One example involves receiving input data as part of a two-way communication session associated with a plurality of bots, accessing confidence scores from the bots. Mapped scores are then generated for the plurality of bots from the confidence scores using a bot score mapper. A selected bot is identified using the mapped scores, and the two-way communication session is facilitated using the selected bot. Further, techniques are provided to track performance of the selected bot and dynamically updated mapping adjustments in the bot score mapper using feedback and machine learning systems.
1. A computer-implemented method comprising:
receiving input data as part of a two-way communication session;
simultaneously providing the input data to a plurality of bots;
simultaneously receiving heterogeneous confidence scores for the plurality of bots, wherein a heterogeneous confidence score corresponds to a bot and is generated using an analysis by the bot of the input data, and indicates an association strength between the input data and the bot;
generating mapped scores for the plurality of bots by calibrating the heterogeneous confidence scores with an associated bot score mapper, wherein mapped scores are standardized confidence scores;
selecting a first bot based on a first mapped score for the first bot having a highest value score of the mapped scores;
comparing the first mapped score for the first bot against a first bot specific threshold value;
rejecting the first mapped score when the first mapped score is below the first bot specific threshold value;
selecting a second bot based on a second mapped score for the second bot having a next highest value score of the mapped scores;
comparing a second mapped score for a second bot against a second bot specific threshold value, wherein the second bot specific threshold value is different than the first bot specific threshold value;
selecting the second bot based on the second bot meeting the second bot specific threshold value;
facilitating the two-way communication session using the second bot; and
recalibrating one or more of the standardized confidence scores of the bot score mappers, based on the selection of the second bot and the two-way communication session, thereby adjusting subsequent bot selection.
2. The computer-implemented method of claim 1 , further comprising processing the input data to identify a bot type associated with the two-way communication session.
3. The computer-implemented method of claim 1 , wherein the plurality of bots are associated with a respective bot type.
4. The computer-implemented method of claim 1 , wherein the selected second bot does not have a highest confidence score of self-generated confidence scores generated by the plurality of bots; and
wherein the selected second bot has a highest mapped score of the mapped scores for the plurality of bots.
5. The computer-implemented method of claim 1 , further comprising:
receiving feedback data; and
updating a mapping adjustment value associated with the input data and the selected second bot of the bot score mapper based on feedback data, wherein the feedback data further includes manual feedback provided by an agent participating in the two-way communication session.
6. The computer-implemented method of claim 1 , further comprising:
receiving an agent input rejecting the selected second bot; and
selecting an additional bot based on the mapped scores excluding a score of the selected bot.
7. The computer-implemented method of claim 1 , wherein the plurality of bots includes one or more client bots, one or more source bots, and one or more external bots.
8. The computer-implemented method of claim 1 , further comprising:
transmitting one or more confidence score requests, wherein when the one or more confidence score requests are received by a server executing one or more bots, the server generates confidence scores for the one or more bots.
9. The computer-implemented method of claim 1 , further comprising:
executing instructions for the plurality of bots using one or more processors of a device; and
generating the mapped scores using the one or more processors of the device.
10. A system comprising:
a memory; and
one or more processors coupled to the memory, the one or more processors configured to perform operations comprising:
receiving input data as part of a two-way communication session;
simultaneously providing the input data to a plurality of bots;
simultaneously receiving heterogeneous confidence scores for the plurality of bots, wherein a heterogeneous confidence score corresponds to a bot and is generated using an analysis by the bot of the input data, and indicates an association strength between the input data and the bot;
generating mapped scores for the plurality of bots by calibrating the heterogeneous confidence scores with an associated bot score mapper, wherein mapped scores are standardized confidence scores;
selecting a first bot based on a first mapped score for the first bot having a highest value score of the mapped scores;
comparing the first mapped score for the first bot against a first bot specific threshold value;
rejecting the first mapped score when the first mapped score is below the first bot specific threshold value;
selecting a second bot based on a second mapped score for the second bot having a next highest value score of the mapped scores
comparing a second mapped score for a second bot against a second bot specific threshold value, wherein the second bot specific threshold value is different than the first bot specific threshold value;
selecting the second bot based on the second bot meeting the second bot specific threshold value;
facilitating the two-way communication session using the second bot; and
recalibrating one or more of the standardized confidence scores of the bot score mappers. based on the selection of the second bot and the two-way communication session, thereby adjusting subsequent bot selection.
11. The system of claim 10 , wherein the one or more processors are configured for operations further comprising processing the input data to identify a bot type associated with the two-way communication session.
12. The system of claim 10 , wherein the plurality of bots are associated with a respective bot type.
13. The system of claim 10 , wherein the selected second bot does not have a highest confidence score of self-generated confidence scores generated by the plurality of bots; and
wherein the selected second bot has a highest mapped score of the mapped scores for the plurality of bots.
14. The system of claim 10 , wherein the one or more processors are configured for operations further comprising:
receiving feedback data; and
updating a mapping adjustment value associated with the input data and the selected second bot of the bot score mapper based on feedback data, wherein the feedback data further includes manual feedback provided by an agent participating in the two-way communication session.
15. The system of claim 10 , wherein the one or more processors are configured for operations further comprising:
receiving an agent input rejecting the selected second bot; and
selecting an additional bot based on the mapped scores excluding a score of the selected bot.
16. The system of claim 10 , wherein the plurality of bots includes one or more client bots, one or more source bots, and one or more external bots.
17. The system of claim 10 , wherein the one or more processors are configured for operations further comprising:
transmitting one or more confidence score requests, wherein when the one or more confidence score requests are received by a server executing one or more bots, the server generates confidence scores for the one or more bots.
18. The system of claim 10 , wherein the one or more processors are configured for operations further comprising:
executing instructions for the plurality of bots using one or more processors of a device; and
generating the mapped scores using the one or more processors of the device.
19. A non-transitory computer readable medium comprising instructions that, when executed by one or more processors of a device, cause the device to perform operations comprising:
receiving input data as part of a two-way communication session;
simultaneously providing the input data to a plurality of bots;
simultaneously receiving heterogeneous confidence scores for the plurality of bots, wherein a heterogeneous confidence score corresponds to a bot and is generated using an analysis by the bot of the input data, and indicates an association strength between the input data and the bot;
generating mapped scores for the plurality of bots by calibrating the heterogeneous confidence scores with an associated bot score mapper, wherein mapped scores are standardized confidence scores;
selecting a first bot based on a first mapped score for the first bot having a highest value score of the mapped scores;
comparing the first mapped score for the first bot against a first bot specific threshold value;
rejecting the first mapped score when the first mapped score is below the first bot specific threshold value;
selecting a second bot based on a second mapped score for the second bot having a next highest value score of the mapped scores
comparing a second mapped score for a second bot against a second bot specific threshold value, wherein the second bot specific threshold value is different than the first bot specific threshold value;
selecting the second bot based on the second bot meeting the second bot specific threshold value;
facilitating the two-way communication session using the second bot; and
recalibrating one or more of the standardized confidence scores of the bot score mappers, based on the selection of the second bot and the two-way communication session, thereby adjusting subsequent bot selection.
20. The non-transitory computer readable medium of claim 19 , wherein when executed by one or more processors of a device, the instructions cause the device to perform operations further comprising processing the input data to identify a bot type associated with the two-way communication session.
21. The non-transitory computer readable medium of claim 19 , wherein the plurality of bots are associated with a respective bot type.
22. The non-transitory computer readable medium of claim 19 , wherein the selected bot does not have a highest confidence score of self-generated confidence scores generated by the plurality of bots; and
wherein the selected bot has a highest mapped score of the mapped scores for the plurality of bots.
23. The non-transitory computer readable medium of claim 19 , wherein when executed by one or more processors of a device, the instructions cause the device to perform operations further comprising:
receiving feedback data; and
updating a mapping adjustment value associated with the input data and the selected bot of the bot score mapper based on feedback data, wherein the feedback data further includes manual feedback provided by an agent participating in the two-way communication session.
24. The non-transitory computer readable medium of claim 19 , wherein when executed by one or more processors of a device, the instructions cause the device to perform operations further comprising:
receiving an agent input rejecting the selected second bot; and
selecting an additional bot based on the mapped scores excluding a score of the selected bot.
25. The non-transitory computer readable medium of claim 19 , wherein the plurality of bots includes one or more client bots, one or more source bots, and one or more external bots.
26. The non-transitory computer readable medium of claim 19 , wherein when executed by one or more processors of a device, the instructions cause the device to perform operations further comprising:
transmitting one or more confidence score requests, wherein when the one or more confidence score requests are received by a server executing one or more bots, the server generates confidence scores for the one or more bots.
27. The non-transitory computer readable medium of claim 19 , wherein when executed by one or more processors of a device, the instructions cause the device to perform operations further comprising:
executing instructions for the plurality of bots using one or more processors of a device; and
generating the mapped scores using the one or more processors of the device.