Dynamic response prediction for improved bot task processing
Systems and methods can be provided for predicting responses during communication sessions with network devices. In some implementations, systems and methods can facilitate predicting responses using machine learning techniques. Messages received through a platform can be stored in a repository. A machine learning model may be trained using the stored messages. When a terminal device is communicating with a network device in a communication session, the messages exchanged in the communication session and the machine learning model can be used to predict future responses in real-time. The predicted future responses can be presented at the terminal device. A predicted response can be selected at the terminal device. Upon selection, the selected predicted response is transmitted to the network device during the communication session.
1. A computer-implemented method, comprising:
facilitating a communication session between a terminal device and a network device, wherein the communication session enables the terminal device and the network device to exchange one or more messages;
receiving a portion of a new message during the communication session;
generating an evaluation of the new message using a machine-learning model and the portion of the new message, wherein the machine-learning model includes a matrix matching model where candidate predicted messages in a message store correspond to matrixes of vector parameters;
continuously performing prediction system queries to one or more databases in real-time as additional portions of the new message are received, wherein the prediction system queries include vector parameter matrix representations of received portions of the new message;
continuously receiving prediction system query responses;
analyzing the prediction system query responses using the matrix matching model to generate updates to the evaluation of the new message continuously in real-time as the additional portions of the new message are received; and
transmitting the updates to the evaluation, wherein when the new message is received from the network device and the updates to the evaluation are transmitted to the terminal device, the terminal device presents the updates to the evaluation in real-time as a predicted response to the new message.
2. The computer-implemented method of claim 1 , further comprising generating a vector parameter matrix for the portion of the new message; and
generating an updated vector parameter matrix based on the additional portions of the new message as they are received.
3. The computer-implemented method of claim 1 ;
wherein the prediction system query responses include responses matched to the vector parameter matrix representations of the prediction system queries.
4. The computer-implemented method of claim 1 , further comprising:
collecting a data set for training the machine-learning model, wherein collecting the data set includes storing previous messages included in previous communication sessions between network devices and terminal devices.
5. The computer-implemented method of claim 1 , further comprising:
collecting agent responses to messages in previous communication sessions between network devices and terminal devices;
generating vector parameter matrixes for the messages; and
storing the vector parameter matrixes and the agent responses in the one or more databases.
6. The computer-implemented method of claim 1 , further comprising:
receiving a selected response associated with the predicted response; and
updating the one or more databases with training data associated with the new message and the selected response.
7. The computer-implemented method of claim 1 , further comprising automatically transmitting the predicted response, wherein when the new message is received from the network device, the predicted response is automatically transmitted to the network device.
8. A system, comprising:
one or more data processors; and
a non-transitory computer-readable storage medium containing instructions which, when executed on the one or more data processors, cause the one or more data processors to perform operations including:
facilitating a communication session between a terminal device and a network device, wherein the communication session enables the terminal device and the network device to exchange one or more messages;
receiving a portion of a new message during the communication session;
generating an evaluation of the new message using a machine-learning model and the portion of the new message, wherein the machine-learning model includes a matrix matching model where candidate predicted messages in a message store correspond to matrixes of vector parameters;
continuously performing prediction system queries to one or more databases in real-time as additional portions of the new message are received, wherein the prediction system queries include vector parameter matrix representations of received portions of the new message;
continuously receiving prediction system query responses;
analyzing the prediction system query responses using the matrix matching model to generate updates to the evaluation of the new message continuously in real-time as the additional portions of the new message are received; and
transmitting the updates to the evaluation, wherein when the new message is received from the network device and the updates to the evaluation are transmitted to the terminal device, the terminal device presents the updates to the evaluation in real-time as a predicted response to the new message.
9. The system of claim 8 , wherein the operations further include:
generating a vector parameter matrix for the portion of the new message; and
generating an updated vector parameter matrix based on the additional portions of the new message as they are received.
10. The system of claim 8 ;
wherein the prediction system query responses include responses matched to the vector parameter matrix representations of the prediction system queries.
11. The system of claim 8 , wherein the operations further include:
collecting a data set for training the machine-learning model, wherein collecting the data set includes storing previous messages included in previous communication sessions between network devices and terminal devices.
12. The system of claim 8 , wherein the operations further include:
collecting agent responses to messages in previous communication sessions between network devices and terminal devices;
generating vector parameter matrixes for the messages; and
storing the vector parameter matrixes and the agent responses in the one or more databases.
13. The system of claim 8 , wherein the operations further include:
receiving a selected response associated with the predicted response; and
updating the one or more databases with training data associated with the new message and the selected response.
14. The system of claim 8 , wherein the operations further include:
automatically transmitting the predicted response, wherein when the new message is received from the network device, the predicted response is automatically transmitted to the network device.
15. A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, including instructions configured to cause a data processing apparatus to perform operations including:
facilitating a communication session between a terminal device and a network device, wherein the communication session enables the terminal device and the network device to exchange one or more messages;
receiving a portion of a new message during the communication session;
generating an evaluation of the new message using a machine-learning model and the portion of the new message, wherein the machine-learning model includes a matrix matching model where candidate predicted messages in a message store correspond to matrixes of vector parameters;
continuously performing prediction system queries to one or more databases in real-time as additional portions of the new message are received, wherein the prediction system queries include vector parameter matrix representations of received portions of the new message;
continuously receiving prediction system query responses;
analyzing the prediction system query responses using the matrix matching model to generate updates to the evaluation of the new message continuously in real-time as the additional portions of the new message are received; and
transmitting the updates to the evaluation, wherein when the new message is received from the network device and the updates to the evaluation are transmitted to the terminal device, the terminal device presents the updates to the evaluation in real-time as a predicted response to the new message.
16. The computer-program product of claim 15 , wherein the operations further include:
generating a vector parameter matrix for the portion of the new message; and
generating an updated vector parameter matrix based on the additional portions of the new message as they are received.
17. The computer-program product of claim 15 ;
wherein the prediction system query responses include responses matched to the vector parameter matrix representations of the prediction system queries.
18. The computer-program product of claim 15 , wherein the operations further include:
collecting a data set for training the machine-learning model, wherein collecting the data set includes storing previous messages included in previous communication sessions between network devices and terminal devices.
19. The computer-program product of claim 15 , wherein the operations further include:
collecting agent responses to messages in previous communication sessions between network devices and terminal devices;
generating vector parameter matrixes for the messages; and
storing the vector parameter matrixes and the agent responses in the one or more databases.
20. The computer-program product of claim 15 , wherein the operations further include:
receiving a selected response associated with the predicted response; and
updating the one or more databases with training data associated with the new message and the selected response.