DESIGN AND IMPLEMENTATION OF A REAL TIME CYBER THREAT DETECTION SYSTEM USING MACHINE LEARNING FOR CRITICAL INFRASTRUCTURE

As cyber threats become increasingly sophisticated , they pose significant risks to critical infrastructure and organizational assets, creating a need for intelligent and proactive cybersecurity solutions. This study presents a Real Time Cyber Threat Detection System (CTDS) that integrates machine learning with big data analytics to improve the speed and accuracy of cyber threat detection. The research adopts a systematic methodology involving the CICIDS2017 dataset selection and preprocessing, model development , system implementation and performance evaluation. The Random Forest machine learning algorithm was applied to classify network traffic and detect malicious activities , while accuracy, precision, recall, and F1-score were used to assess system performance . The proposed system enables continuous real-time network monitoring , allowing organizations to identify and respond to potential cyber attacks promptly. Experimental results shows that Random Forest model achieved an accuracy of 99.59%, precision of 99.85%, recall of 99.59%, and F1-score of 99.69% , effectively differentiating between normal and malicious network traffic. The study also highlights challenges associated with real- time deployment, including data scalability, and suggests future improvements to address these limitations. Overall, this research contributes to the field of cyber threat intelligence by providing a practical framework for integrating machine learning into real-time threat detection . The findings emphasize the importance of predictive analytics, anomaly detection, and proactive defense mechanism in strengthening cybersecurity and protecting critical infrastructure.