Systems and methods of gateway detection in a telephone network
Embodiments described herein provide for detecting whether an Automatic Number Identification (ANI) associated with an incoming call is a gateway, according to rules-based models and machine learning models generated by the computer using call data stored in one or more databases.
1. A computer-implemented method for detecting gateways, the method comprising:
obtaining, by a computer, call data of a plurality of prior calls associated with a plurality of automatic number identifications (ANIs);
for each ANI of the plurality of ANIs:
extracting, by the computer, a plurality of features from one or more prior calls associated with the ANI of the plurality of prior calls;
generating, by the computer, a feature vector for the ANI based upon the plurality features extracted from the one or more prior calls associated with the ANI; and
generating, by the computer, a gateway score for the ANI profile by applying an anomaly detection to the ANI;
obtaining, by the computer, a gateway threshold and a contamination factor associated with the gateway threshold; and
identifying, by the computer, at least one ANI as at least one gateway based upon the gateway score of the at least one ANI and the gateway threshold.
2. The method according to claim 1 , further comprising determining, by the computer, a number of gateways identified by the computer in the at least one gateway.
3. The method according to claim 1 , further comprising adjusting, by the computer, the gateway threshold according to the contamination factor and a number of gateways identified by the computer in the at least one gateway.
4. The method according to claim 1 , further comprising:
generating, by the computer, an inbound feature vector for an inbound ANI based upon a plurality inbound features extracted from inbound call data of an inbound call; and
generating, by the computer, an inbound gateway score for the inbound ANI based upon the inbound feature vector.
5. The method according to claim 4 , further comprising identifying, by the computer, the inbound ANI as a gateway based upon the inbound gateway score and the gateway threshold.
6. The method according to claim 4 , further comprising authenticating, by the computer, the inbound call associated with the inbound ANI based upon the inbound gateway score the gateway threshold.
7. The method according to claim 1 , wherein obtaining the contamination factor includes receiving, by the computer, the contamination factor from a client device.
8. The method according to claim 1 , wherein the plurality of features include at least one of:
a number of calls from the ANI to each of the callees receiving a comparatively larger number of calls normalized by a calls threshold,
a number of calls from the ANI to each type of callee receiving the comparatively larger number of calls,
a lowest interarrival interval between consecutive calls, or
a network type of the ANI.
9. The method according to claim 1 , wherein the computer obtains the call data and feature vector for each ANI according to a lookback window.
10. The method according to claim 1 , wherein obtaining the call data of the plurality of prior calls comprises, for each ANI:
generating, by the computer, a volume score, a spread score, and a call diversity score;
generating, by the computer, a rules-based gateway score; and
generating, by the computer, a combined gateway score based upon the gateway score generated using the feature vector and the rules-based gateway score.
11. A system comprising:
a computer comprising a processor and non-transitory storage medium containing executable instructions, and configured to:
obtain call data for a plurality of prior calls associated with a plurality of automatic number identifications (ANIs);
for each ANI of the plurality of ANIs:
extract a plurality of features from one or more prior calls associated with the ANI of the plurality of prior calls;
generate a feature vector for the ANI based upon the plurality features extracted from the one or more prior calls associated with the ANI; and
generate a gateway score for the ANI profile by applying an anomaly detection to the ANI;
obtain a gateway threshold and a contamination factor associated with the gateway threshold; and
identify at least one ANI as at least one gateway based upon the gateway score of the at least one ANI and the gateway threshold.
12. The system according to claim 11 , wherein the computer is further configured to determine a number of gateways identified by the computer in the at least one gateway.
13. The system according to claim 11 , wherein the computer is further configured to adjust the gateway threshold according to the contamination factor and a number of gateways identified by the computer in the at least one gateway.
14. The system according to claim 11 , wherein the computer is further configured to:
generate an inbound feature vector for an inbound ANI based upon a plurality inbound features extracted from inbound call data of an inbound call; and
generate an inbound gateway score for the inbound ANI based upon the inbound feature vector.
15. The system according to claim 14 , wherein the computer is further configured to identify the inbound ANI as a gateway based upon the inbound gateway score and the gateway threshold.
16. The system according to claim 14 , wherein the computer is further configured to authenticate the inbound call associated with the inbound ANI based upon the inbound gateway score the gateway threshold.
17. The system according to claim 11 , wherein when obtaining the contamination factor wherein the computer is further configured to receive the contamination factor from a client device.
18. The system according to claim 11 , wherein the plurality of features include at least one of:
a number of calls from the ANI to each of the callees receiving a comparatively larger number of calls normalized by a calls threshold,
a number of calls from the ANI to each type of callee receiving the comparatively larger number of calls,
a lowest interarrival interval between consecutive calls, or
a network type of the ANI.
19. The system according to claim 11 , wherein the computer is configured to obtain the call data and feature vector for each ANI according to a lookback window.
20. The system according to claim 11 , wherein when obtaining the call data of the plurality of prior calls the computer is further configured to, for each ANI:
generate a volume score, a spread score, and a call diversity score;
generate a rules-based gateway score; and
generate a combined gateway score based upon the gateway score generated using the feature vector and the rules-based gateway score.