Method and system for training a virtual agent using optimal utterances
A method and system for training a virtual agent is provided herein. The method and system comprises storing conversations between the virtual agent and a user in logs. The method and system further comprises mining the logs to retrieve utterances. The method and system further comprises computing regression for each of the plurality of the charging time segments. The method and system further comprises providing a score to the utterances. Further, the method ranking the utterances based on the score.
1 . A computer-implemented method for training a virtual agent, comprising:
storing conversations between the virtual agent and a user in logs;
mining the logs to retrieve utterances;
assigning a score to each of the utterances;
ranking the utterances based on the assigned scores; and
determining one or more of the ranked utterances as optimal based on a threshold score wherein the determining the one or more of the ranked utterances as optimal based on the threshold score comprises marking utterances with scores greater than a first threshold and lesser than a second threshold as optimal.
2 . The computer-implemented method of claim 1 , wherein the score is calculated based on syntactic similarity.
3 . The computer-implemented method of claim 1 , wherein the score is calculated based on semantic similarity.
4 . The computer-implemented method of claim 3 , further comprising calculating the semantic similarity using a deep learning model.
5 . The computer-implemented method of claim 1 , wherein the score is calculated based on word error rate.
6 . The method of claim 1 , wherein the score is calculated based on language code for a multilingual chat.
7 . The computer-implemented method of claim 1 , further comprising extracting the one or more optimal utterances based on the ranking the utterances.
8 . The computer-implemented method of claim 7 , further comprising training the virtual agent based on the one or more optimal utterances.
9 . A computer system for training a virtual agent, the computer system comprising:
one or more computer processors, one or more computer readable memories, one or more computer readable storage devices, and program instructions stored on the one or more computer readable storage devices for execution by the one or more computer processors via the one or more computer readable memories, the program instructions comprising:
storing conversations between the virtual agent and a user in logs;
mining the logs to retrieve utterances;
assigning a score to each of the utterances;
ranking the utterances based on the assigned scores; and
determining one or more of the ranked utterances as optimal based on a threshold score wherein the determining the one or more of the ranked utterances as optimal based on the threshold score comprises marking utterances with scores greater than a first threshold and lesser than a second threshold as optimal.
10 . The computer system of claim 9 , wherein the score is calculated based syntactic similarity.
11 . The computer system of claim 9 , wherein the score is calculated based on semantic similarity.
12 . The computer system of claim 11 , further comprising calculating the semantic similarity using a deep learning model.
13 . The computer system of claim 9 , wherein the score is calculated based on word error rate.
14 . The computer system of claim 9 , wherein the score is calculated based on language code for a multilingual chat.
15 . The computer system of claim 9 , further comprising extracting the one or more optimal utterances based on the ranking the utterances.
16 . The computer system of claim 15 , further comprising training the virtual agent based on the one or more optimal utterances.
17 . A non-transitory computer-readable storage medium having stored thereon computer executable instructions which, when executed by one or more processors, cause the one or more processors to carry out operations for training a virtual agent, the operations comprising:
storing conversations between the virtual agent and a user in logs;
mining the logs to retrieve utterances;
assigning a score to each of the utterances;
ranking the utterances based on the assigned scores; and
determining one or more of the ranked utterances as optimal based on a threshold score wherein the determining the one or more of the ranked utterances as optimal based on the threshold score comprises marking utterances with scores greater than a first threshold and lesser than a second threshold as optimal.