DESIGN OF PREDICTIVE COUNTER TERRORISM SYSTEM

Terrorism remains one of the most serious threats to global peace and national security, with its devastating impacts felt across societies, economies, and governance systems. Terrorist acts are often unpredictable, strategically planned, and executed in a manner that avoids early detection by conventional security and surveillance systems. Traditional counter-terrorism methods have largely depended on manual intelligence gathering, human surveillance, and post-incident analysis, which in many situations exposed them to delays, human errors, and failed attempts to address threats that are more complex or coordinated in real time. Considering the sophistication that terrorist networks continue to acquire and the changing nature of these threats, there is a pressing need for a more proactive and intelligent approach to security. This research has developed a predictive counter-terrorism system, aimed at detecting potential threats at a very early stage, analyzing risk factors from possible sources, and offering actionable intelligence for timely preemptive intervention. The project relied on Random Forest for prediction and HTML for creating the web interface. Flask was used to create interactive graphs, and Python was used for coding. The system was tested for validity and deployed for web use. It enabled accurate detection and prediction of terrorist activities before execution, providing much-needed efficacy in national defense, allowing security agencies to neutralize threats with minimal disruption.