Machine learning dataset generation using a natural language processing technique
A server can receive a plurality of records at a databases such that each record is associated with a phone call and includes at least one request generated based on a transcript of the phone call. The server can generate a training dataset based on the plurality of records. The server can further train a binary classification model using the training dataset. Next, the server can receive a live transcript of a phone call in progress. The server can generate at least one live request based on the live transcript using a natural language processing module of the server. The server can provide the at least one live request to the binary classification model as input to generate a prediction. Lastly, the server can transmit the prediction to an entity receiving the phone call in progress. The prediction can cause a transfer of the call to a chatbot.
1. A method, comprising:
receiving, at a server, a plurality of records, wherein each record includes a call recording, a phone number, and a fraud designation being one of fraudulent or non-fraudulent;
creating, using a processor, a training dataset including a plurality of data points, each data point including the call recording, the phone number, and the fraud designation;
training, using the processor, a classification model using the training dataset;
receiving, at the server, a new phone call from a new phone number;
labeling, using the processor, the new phone call with a new fraud designation based on a classification by the classification model, wherein the new fraud designation is fraudulent or non-fraudulent; and
transmitting, using the processor, a transfer signal to a device to transfer the new phone call when the new fraud designation is fraudulent.
2. The method of claim 1 , wherein the transfer signal is configured to cause a transfer of the new phone call to a chatbot.
3. The method of claim 2 , wherein the chatbot is configured to receive a verbal query from a caller and provide a response to the caller based on the verbal query.
4. The method of claim 3 , wherein the response is a verbal response, a delay in response or a placement of the new phone call on hold.
5. The method of claim 1 , wherein the transfer signal is configured to cause a transfer of the new phone call to an Interactive Voice Response (IVR) phone loop.
6. The method of claim 1 , wherein each record includes an account number or a time of the new phone call.
7. The method of claim 1 further comprising detecting a background noise for each record.
8. The method of claim 7 further comprising adding the background noise for each record to a data point for the record.
9. The method of claim 1 further comprising detecting a voice profile for a caller for each record.
10. The method of claim 9 further comprising adding the voice profile for each record to a data point for the record.
11. The method of claim 1 further comprising determining, using a natural language processing module, an intent of the call recording.
12. The method of claim 11 wherein the intent is determined based on keywords used in a conversation in the call recording.
13. The method of claim 11 wherein the intent is determined using a machine learning model.
14. The method of claim 13 further comprising using an intent recognition module that includes preprocessing modules to convert text into character, word, or sentence embeddings that is fed into the machine learning model.
15. The method of claim 1 further comprising generating a new call recording or detecting a new background noise for the new phone call.
16. The method of claim 15 further comprising labeling the new phone call using the new call recording or the new background noise.
17. The method of claim 15 further comprising comparing the new background noise to a plurality of known background noises.
18. The method of claim 17 further comprising providing the new phone number to the classification model if the new background noise matches a known background noise.
19. The method of claim 1 , wherein the training dataset is generated by sampling a dataset using a sampling technique.
20. The method of claim 19 , wherein the sampling technique is undersampling the dataset or oversampling the dataset.
21. The method of claim 19 , wherein the sampling technique is a Synthetic Minority Over-sampling Technique, a Modified synthetic minority oversampling technique, a Random Under-Sampling, or a Random Over-Sampling.