Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{179829,
        author = {Bokka Sudheer and Uday Bhaskar and Uday Kumar Reddy Y and Shabaz K and Ms. Soumya G D},
        title = {DDos Protection System using Machine Learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {12},
        pages = {8055-8064},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=179829},
        abstract = {Distributed Denial of Service (DDoS) 
attacks pose one of the greatest threats to the 
availability and reliability of Internet-based services. 
Target systems are flooded with an enormous amount 
of malicious traffic, making it impossible for legitimate 
users to gain access. Rule-based and signature-based 
detection mechanisms are usually ineffective when 
dealing with sophisticated or dynamic DDoS attacks, 
particularly those that resemble normal traffic flow. 
To overcome these limitations, this paper introduces a 
machine learning-based DDoS detection and 
protection system that is capable of automatically 
detecting and blocking abnormal traffic behaviors in 
real-time. 
The system to be implemented utilizes supervised 
learning algorithms that are trained on actual traffic 
datasets like CICIDS2017 and NSL-KDD to identify 
different types of DDoS attacks like SYN flood, UDP 
flood, HTTP GET flood, and ICMP- based attacks. 
The system design has several components such as 
traffic sniffing, feature extraction, model training, and 
an automated mitigation engine. Properties such as 
packet length, source IP entropy, inter-arrival time, 
and protocol type are retrieved and input to a trained 
classifier (e.g., Random Forest or SVM) for attack 
identification. 
The model is highly accurate in identifying malicious 
traffic versus legitimate requests and can be used in 
real-time settings with low overhead. In response to an 
attack detection, the system may automatically block 
suspicious IPs or call upon alert mechanisms for 
administrators. Through this work, the capability of 
machine learning methods to enhance dramatically the 
detection and prevention of DDoS attacks is shown, 
leading to more secure and robust cloud and network 
infrastructures.},
        keywords = {Cloud Security, DDoS Protection, Traffic  Monitoring, Network Forensics, Snort, iptables,  FastAPI.},
        month = {May},
        }