Authors:
Surya Lokesh Bhargav Pentakota

Addresses:
Department of Research and Development, Ginger Labs, Texas, United States of America.

Abstract:

One of the most dangerous types of attacks is known as a distributed denial of service attack (DDoS). In most cases, traditional defensive techniques are unable to identify or prevent innovative, high-speed, or advanced attacks. A real-time artificial intelligence-based monitoring system is proposed in this study for the early detection and prevention of distributed denial-of-service attacks. Identifying attacks is accomplished through the utilisation of a hybrid approach that combines a random forest classifier with a deep neural network-based feature extraction. This system has been trained and tested on the CIC-DDoS2019 dataset, a comprehensive collection of modern distributed denial-of-service attacks on traffic in the modern world. During the system's implementation, network traffic is monitored in real-time, and significant features are extracted to differentiate between malicious and normal packets. The test results demonstrate that the system has a high level of accuracy, a low rate of false positives, and a fast response rate, all of which indicate its effectiveness in real-world settings. Matplotlib is used for data visualisation, whereas Python, TensorFlow, Scikit-learn, and Pandas are research tools that serve as libraries for data handling and model construction.  

Keywords: Artificial Intelligence; Machine Learning; Real-Time Monitoring; Network Security; Distributed Denial of Service (DDoS); Random Forest Classifier; DDoS Attacks; Cybersecurity Frameworks.

Received on: 02/11/2024, Revised on: 07/01/2025, Accepted on: 15/02/2025, Published on: 05/09/2025

DOI: 10.69888/FTSCL.2025.000427

FMDB Transactions on Sustainable Computer Letters, 2025 Vol. 3 No. 3, Pages: 116-125

• Views : 113
• Downloads : 4

Download PDF