Robocall detection
A method of detecting robocalls includes generating at least one of an identification (ID) spoofing score, a number similarity score, and a call length prediction request for a telephone call request. The method further includes generating a robocall predictor score based at least in part on the call length prediction score and the ID spoofing score, and rejecting the telephone call request when the robocall predictor score indicates the telephone call request is invalid.
1. An apparatus comprising:
a processing device; and
a memory device coupled to the processing device, the memory device having instructions stored thereon that, in response to execution by the processing device, cause the processing device to:
generate at least one of an identification (ID) spoofing score and a number similarity score for a telephone call request;
generate a call length prediction score for the telephone call request based at least in part on the at least one of the ID spoofing score and the number similarity score;
generate a robocall predictor score based at least in part on the call length prediction score and the ID spoofing score; and
reject the call request when the robocall predictor score indicates the telephone call request is invalid.
2. The apparatus of claim 1 , wherein the instructions, when executed, cause the processing device to communicate a random number to a caller of the telephone call request and require the caller to input the random number using the caller's telephone keypad or the telephone caller request is rejected.
3. The apparatus of claim 1 , wherein the instructions, when executed, cause the processing device to generate the at least one of the ID spoofing score and the number similarity score for the telephone call request using an ID spoofing machine learning model.
4. The apparatus of claim 3 , wherein the instructions, when executed, cause the processing device to generate the ID spoofing score from one or more of the number similarity score, a day of a week of the call request, and a time of day of the telephone call request.
5. The apparatus of claim 3 , wherein the instructions, when executed, cause the processing device to generate the call length prediction score for the telephone call request using a call length machine learning model.
6. The apparatus of claim 5 , wherein the instructions, when executed, cause the processing device to generate the call length prediction score from one or more of the number similarity score, the ID spoofing score, a day of a week of the telephone call request, a time of day of the telephone call request, and a geographic information of a caller of the telephone call request.
7. The apparatus of claim 5 , wherein the instructions, when executed, cause the processing device to store information about the telephone call request in a database and periodically train the ID spoofing machine learning model and the call length machine learning model using the telephone call request information in the database.
8. The apparatus of claim 1 , wherein the robocall predictor score is a probability that the telephone call request is for a robocall.
9. The apparatus of claim 8 , wherein the instructions, when executed, cause the processing device to reject the telephone call request when the robocall predictor score meets or exceeds a predetermined threshold.
10. The apparatus of claim 1 , wherein the instructions, when executed, cause the processing device to transfer the telephone call request back to the caller, instead of rejecting the telephone call request, when the robocall predictor score indicates the telephone call request is invalid.
11. A method comprising:
generating at least one of an identification (ID) spoofing score and a number similarity score for a telephone call request;
generating a call length prediction score for the telephone call request based at least in part on the at least one of the ID spoofing score and the number similarity score;
generating a robocall predictor score based at least in part on the call length prediction score and the ID spoofing score; and
rejecting the telephone call request when the robocall predictor score indicates the call request is invalid.
12. The method of claim 11 , comprising communicating a random number to a caller of the telephone call request and requiring the caller to input the random number using the caller's telephone keypad or the telephone caller request is rejected.
13. The method of claim 11 , comprising generating the at least one of the ID spoofing score and the number similarity score for the telephone call request using an ID spoofing machine learning model.
14. The method of claim 13 , comprising generate the ID spoofing score from one or more of the number similarity score, a day of a week of the telephone call request, and a time of day of the telephone call request.
15. The method of claim 13 , comprising generating the call length prediction score for the telephone call request using a call length machine learning model.
16. The method of claim 15 , comprising generating the call length prediction score from one or more of the number similarity score, the ID spoofing score, a day of a week of the telephone call request, a time of day of the telephone call request, and a geographic information of a caller of the telephone call request.
17. At least one non-transitory machine-readable storage medium comprising instructions that, when executed, cause at least one processor to:
generate at least one of an identification (ID) spoofing score and a number similarity score for a telephone call request;
generate a call length prediction score for the telephone call request based at least in part on the at least one of the ID spoofing score and the number similarity score;
generate a robocall predictor score based at least in part on the call length prediction score and the ID spoofing score; and
reject the call request when the robocall predictor score indicates the telephone call request is invalid.
18. The least one non-transitory machine-readable storage medium of claim 17 comprising instructions that, when executed, cause at least one processor to generate the at least one of the ID spoofing score and the number similarity score for the telephone call request using an ID spoofing machine learning model.
19. The least one non-transitory machine-readable storage medium of claim 18 comprising instructions that, when executed, cause at least one processor to generate the call length prediction score for the telephone call request using a call length machine learning model.
20. The least one non-transitory machine-readable storage medium of claim 19 comprising instructions that, when executed, cause at least one processor to store information about the telephone call request in a database and periodically train the ID spoofing machine learning model and the call length machine learning model using the telephone call request information in the database.