Penerapan Sistem Keamanan Jaringan Menggunakan Intrusion Prevention System Dengan Machine Learning

Technological advancements have increased dependence on computer networks across various sectors but have also opened vulnerabilities to cyberattacks such as Syn Flood, Port Scanning, and SSH Patator. These attacks can cause service disruptions, data theft, and even system damage. To address this issue, this research aims to design and implement an Intrusion Prevention System (IPS) based on Machine Learning using the Random Forest algorithm. The system is capable of automatically detecting and preventing attacks based on models trained with the CICIDS2017 and CICIDS2019 datasets. In addition, the system is equipped with a mechanism to automatically block the attacker's Internet Protocol (IP) address using Iptables and sends real-time notifications via Telegram to provide a faster response to network administrators. The testing results show that the Random Forest model achieved an accuracy of 97.2% with an F1-score of 95.5%. The system also successfully blocked malicious IP addresses automatically using Iptables. Therefore, the proposed system enhances network security by providing more accurate attack detection and efficient automated response without the need for continuous manual monitoring.