Systems and methods for transferring messaging to automation
The present disclosure relates generally to facilitating routing of communications. More specifically, techniques are provided to dynamically transfer messaging between a network device and a terminal device to a type of bot based on intents identified from the messaging. Further, techniques are provided to track performance of the selected type of bot during automation.
1. A computer-implemented method, comprising:
determining an intent associated with a conversation, wherein messages are exchanged in real-time during the conversation, and wherein the conversation is associated with a terminal device;
receiving the messages in real-time as the messages are exchanged;
using the intent and the messages as input to a trained machine learning model to generate a recommended response message, wherein the trained machine learning model identifies a cluster of messages based on the intent, and wherein the recommended response message is generated as a result of a confidence score associated with the recommended response message satisfying a confidence threshold;
providing the recommended response message, wherein when the recommended response is received at the terminal device, an agent associated with the terminal device determines whether to communicate the recommended response message or an alternative response message;
dynamically determining feedback corresponding to an actual response message exchanged during the conversation, wherein the actual response message includes either the recommended response message or the alternative response message;
determining a polarity for the conversation based on the feedback; and
updating the trained machine learning model using the intent, the actual response message, and the polarity to determine new recommended response messages, wherein the trained machine learning model is updated by modifying the identified cluster of messages according to the actual response message, the polarity, and the feedback.
2. The computer-implemented method of claim 1 , further comprising:
recording messages corresponding to the agent during the conversation; and
associating the messages corresponding to the agent with the intent based on the polarity for the conversation.
3. The computer-implemented method of claim 1 , wherein the terminal device is selected based on a likelihood of the terminal device to provide a positive experience during the conversation.
4. The computer-implemented method of claim 1 , wherein the feedback corresponding to the conversation includes textual and non-textual attributes associated with the messages exchanged during the conversation.
5. The computer-implemented method of claim 1 , wherein the intent is associated with a request for the conversation, and wherein the request is in a natural language.
6. The computer-implemented method of claim 1 , wherein the polarity for the conversation is determined using another trained machine learning model, and wherein the other trained machine learning model is trained to determine the polarity as the messages are exchanged in the conversation.
7. The computer-implemented method of claim 1 , wherein the feedback is dynamically determined by monitoring the conversation to identify a reaction to the actual response message, and wherein the reaction is used to determine the polarity for the conversation.
8. A system, comprising:
one or more processors; and
memory storing thereon instructions that, as a result of being executed by the one or more processors, cause the system to:
determine an intent associated with a conversation, wherein messages are exchanged in real-time during the conversation, and wherein the conversation is associated with a terminal device;
receive the messages in real-time as the messages are exchanged;
use the intent and the messages as input to a trained machine learning model to generate a recommended response message, wherein the trained machine learning model identifies a cluster of messages based on the intent, and wherein the recommended response message is generated as a result of a confidence score associated with the recommended response message satisfying a confidence threshold;
provide the recommended response message, wherein when the recommended response is received at the terminal device, an agent associated with the terminal device determines whether to communicate the recommended response message or an alternative response message;
dynamically determine feedback corresponding to an actual response message exchanged during the conversation, wherein the actual response message includes either the recommended response message or the alternative response message;
determine a polarity for the conversation based on the feedback; and
update the trained machine learning model using the intent, the actual response message, and the polarity to determine new recommended response messages, wherein the trained machine learning model is updated by modifying the identified cluster of messages according to the actual response message, the polarity, and the feedback.
9. The system of claim 8 , wherein the instructions further cause the system to:
record messages corresponding to the agent during the conversation; and
associate the messages corresponding to the agent with the intent based on the polarity for the conversation.
10. The system of claim 8 , wherein the terminal device is selected based on a likelihood of the terminal device to provide a positive experience during the conversation.
11. The system of claim 8 , wherein the feedback corresponding to the conversation includes textual and non-textual attributes associated with the messages exchanged during the conversation.
12. The system of claim 8 , wherein the intent is associated with a request for the conversation, and wherein the request is in a natural language.
13. The system of claim 8 , wherein the polarity for the conversation is determined using another trained machine learning model, and wherein the other trained machine learning model is trained to determine the polarity as the messages are exchanged in the conversation.
14. The system of claim 8 , wherein the feedback is dynamically determined by monitoring the conversation to identify a reaction to the actual response message, and wherein the reaction is used to determine the polarity for the conversation.
15. A non-transitory, computer-readable storage medium storing thereon executable instructions that, as a result of being executed by one or more processors of a computer system, cause the computer system to:
determine an intent associated with a conversation, wherein messages are exchanged in real-time during the conversation, and wherein the conversation is associated with a terminal device;
receive the messages in real-time as the messages are exchanged;
use the intent and the messages as input to a trained machine learning model to generate a recommended response message, wherein the trained machine learning model identifies a cluster of messages based on the intent, and wherein the recommended response message is generated as a result of a confidence score associated with the recommended response message satisfying a confidence threshold;
provide the recommended response message, wherein when the recommended response is received at the terminal device, an agent associated with the terminal device determines whether to communicate the recommended response message or an alternative response message;
dynamically determine feedback corresponding to an actual response message exchanged during the conversation, wherein the actual response message includes either the recommended response message or the alternative response message;
determine a polarity for the conversation based on the feedback; and
update the trained machine learning model using the intent, the actual response message, and the polarity to determine new recommended response messages, wherein the trained machine learning model is updated by modifying the identified cluster of messages according to the actual response message, the polarity, and the feedback.
16. The non-transitory, computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:
record messages corresponding to the agent during the conversation; and
associate the messages corresponding to the agent with the intent based on the polarity for the conversation.
17. The non-transitory, computer-readable storage medium of claim 15 , wherein the terminal device is selected based on a likelihood of the terminal device to provide a positive experience during the conversation.
18. The non-transitory, computer-readable storage medium of claim 15 , wherein the feedback corresponding to the conversation includes textual and non-textual attributes associated with the messages exchanged during the conversation.
19. The non-transitory, computer-readable storage medium of claim 15 , wherein the intent is associated with a request for the conversation, and wherein the request is in a natural language.
20. The non-transitory, computer-readable storage medium of claim 15 , wherein the polarity for the conversation is determined using another trained machine learning model, and wherein the other trained machine learning model is trained to determine the polarity as the messages are exchanged in the conversation.
21. The non-transitory, computer-readable storage medium of claim 15 , wherein the feedback is dynamically determined by monitoring the conversation to identify a reaction to the actual response message, and wherein the reaction is used to determine the polarity for the conversation.