PENGEMBANGAN MODEL DECISION TREE UNTUK SERANGAN DISTRIBUTED DENIAL OF SERVICE

Abstract

In the growing digital age, distributed denial of service (DDoS) attacks have become one of the most pressing and destructive cyber security threats. To address this, the research developed and implemented the Decision Tree model to detect DDoS attacks effectively. An intrusion detection system integrates the model, utilizing machine learning technology to analyze TCP data flows in real-time, with the aim of enhancing network detection capabilities and bolstering security measures against DDoS attacks. We built the Decision Tree model using the NF-UQ-NIDS dataset, which includes network traffic data representative of both DDoS attacks and normal traffic. Early data analysis using Wireshark provides additional insight into attack patterns, which helps with model calibration and validation. The developed system effectively identified attacks and sent real-time notifications via Telegram, facilitating prompt action from the security team. The results of this study show that the integration of machine learning into network security systems offers a significant improvement in the speed and accuracy of attack detection, showing enormous potential for further applications in a dynamic and diverse environment. Recommendations for further research include developing hybrid algorithms, implementing automated responses, and expanding notification platforms to strengthen the cyber security architecture against DDoS attacks and similar threats.