System and method for detecting spam over internet telephony (SPIT) in IP telecommunication systems
A method and system for detecting a spam over internet telephony (SPIT) is disclosed. Incoming call requests are received and analyzed using heuristic algorithms. A runtime action is provided based on the analysis using the heuristic algorithms. The heuristic algorithms include zero-touch heuristic algorithms that require no human intervention, one-touch heuristic algorithms that require human intervention one time, and multi-touch heuristic algorithms that require human intervention more than one time. Cumulative probability is calculated for each incoming call request, and the incoming call request is classified as a spam if the cumulative probability exceeds a threshold.
1. A computer-implemented method, comprising:
receiving an incoming call request;
analyzing the incoming call request using heuristic algorithms, wherein said heuristic algorithms include zero-touch heuristic algorithms that require no human intervention, one-touch heuristic algorithms that require human intervention one time, and multi-touch heuristic algorithms that require human intervention more than one time;
providing a runtime action based on the analysis using said heuristic algorithms;
detecting a spam over internet telephony (SPIT);
calculating cumulative probability for the incoming call request, wherein the cumulative probability adds the probability of the zero-touch heuristic algorithms, the one-touch heuristic algorithms, and the multi-touch heuristic algorithms; and
classifying the incoming call request as a spam when the cumulative probability exceeds a threshold.
2. The computer-implemented method of claim 1 , wherein the zero-touch heuristic algorithms include call duration tracking (CDT), detection of machine-generated requests (DMR), IP/domain correlation (IPD-C), error response learning (ERL), simultaneous calls detection (SCD), call rate tracking (CRT), and call bombing detection (CBD).
3. The computer-implemented method of claim 1 , wherein the multi-touch heuristic algorithms include feedback learning engine (FLE).
4. The computer-implemented method of claim 1 , wherein the threshold is determined by the call profile and the call category.
5. The computer-implemented method of claim 1 , wherein said incoming call request is analyzed in reference to a database that stores list of calls and users.
6. The computer-implemented method of claim 5 , wherein the database classifies the stored calls and users into a whitelist, a blacklist, and a graylist, wherein the whitelist contains the calls and users with legitimate , the blacklist contains the calls and users with malignant intention, and the graylist contains the calls and users that are listed neither in the whitelist nor the blacklist.
7. The computer-implemented method of claim 5 further comprising:
relisting said incoming call request from one list of the whitelist, the blacklist, and the graylist to another list of the whitelist, the blacklist, and the graylist.
8. The computer-implemented method of claim 1 , wherein the runtime action is provided by a remediation engine.
9. The computer-implemented method of claim 8 , wherein the runtime action is one of accept, drop, redirect, temporary blacklist, limited number of message, and alert notification.
10. A system, comprising:
a processor; and
software instructions executable by the processor, the software instructions for configuring the processor to provide
a runtime engine that receives an incoming call request and analyzes the incoming call request using heuristic algorithms; and
a remediation engine that provides a runtime action to the incoming call request, wherein the runtime action determines whether the incoming call request is a spam over internet telephony (SPIT);
wherein said heuristic algorithms include zero-touch heuristic algorithms that require no human intervention, one-touch heuristic algorithms that require human intervention one time, and multi-touch heuristic algorithms that require human intervention more than one time; and
wherein said runtime engine calculates cumulative probability for the incoming call request, wherein the cumulative probability adds the probability of the zero-touch heuristic algorithms, the one-touch heuristic algorithms, and the multi-touch heuristic algorithms and classifies the incoming call request as a spam when the cumulative probability exceeds a threshold.
11. The system of claim 10 , wherein the zero-touch heuristic algorithms include call duration tracking (CDT), detection of machine-generated requests (DMR), IP/domain correlation (IPD-C), error response learning (ERL), simultaneous calls detection (SCD), call rate tracking (CRT), and call bombing detection (CBD).
12. The system of claim 10 , wherein the multi-touch heuristic algorithms include feedback learning engine (FLE).
13. The system of claim 10 , wherein the threshold is determined by the call profile and the call category.
14. The system of claim 10 , further comprising a database that stores list of calls and users, wherein said incoming call request is analyzed in reference to the database.
15. The system of claim 14 , wherein the database classifies the stored calls and users into a whitelist, a blacklist, and a graylist, wherein the whitelist contains the calls and users with legitimate , the blacklist contains the calls and users with malignant intention, and the graylist contains the calls and users that are listed neither in the whitelist nor the blacklist.
16. The system of claim 14 , wherein said incoming call request is relisted from one list of the whitelist, the blacklist, and the graylist to another list of the whitelist, the blacklist, and the graylist.
17. The system of claim 16 , wherein the runtime action is one of accept, drop, redirect, temporary blacklist, limited number of message, and alert notification.