ML – Driven Real Time Classification And Prediction Of Ddos Attacks

  • Unique Paper ID: 195791
  • Volume: 12
  • Issue: 11
  • PageNo: 1399-1403
  • Abstract:
  • Network threats keep changing fast, moving quicker than old defenses can handle. When hackers shift tactics, rule-based firewalls just stall – leaving weird gaps that drive your nuts. Better protection comes from real-time tracking powered by machine learning instead of rigid lists. This system follows traffic patterns, notices tiny irregularities, then dives deep to spot sudden DDoS surges, sneaky scans, erratic probes, early breach attempts, or even unknown moves no one’s seen before. A scoring bit triggers when traffic spikes, links pile up quick, or things shift oddly enough to feel off. It shoots back a threat level – Low, Medium, High – in almost no time at all. Not stuck on Random Forest – it’s okay – but XGBoost runs ahead; my team saw close to 92% accuracy on UNSW-NB15, which caught me off guard more than expected. Python using Flask shows what’s happening right away, flashing alerts directly into view so problems can’t slip through. Instead of replacing old defenses, it moves around them like a shifting layer – always adjusting, spotting changes in traffic the moment they appear, pause hitting on risks before they take hold. It matches today’s pace: quick, responsive, built from real-time data instead of outdated assumptions

Copyright & License

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.

BibTeX

@article{195791,
        author = {Vaishnavi Samaleti and Y. Ashok Kumar and S. Sai Nikhil and D. Lokesh Reddy},
        title = {ML – Driven Real Time Classification And Prediction Of Ddos Attacks},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {1399-1403},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=195791},
        abstract = {Network threats keep changing fast, moving quicker than old defenses can handle. When hackers shift tactics, rule-based firewalls just stall – leaving weird gaps that drive your nuts. Better protection comes from real-time tracking powered by machine learning instead of rigid lists. This system follows traffic patterns, notices tiny irregularities, then dives deep to spot sudden DDoS surges, sneaky scans, erratic probes, early breach attempts, or even unknown moves no one’s seen before. A scoring bit triggers when traffic spikes, links pile up quick, or things shift oddly enough to feel off. It shoots back a threat level – Low, Medium, High – in almost no time at all. Not stuck on Random Forest – it’s okay – but XGBoost runs ahead; my team saw close to 92% accuracy on UNSW-NB15, which caught me off guard more than expected. Python using Flask shows what’s happening right away, flashing alerts directly into view so problems can’t slip through. Instead of replacing old defenses, it moves around them like a shifting layer – always adjusting, spotting changes in traffic the moment they appear, pause hitting on risks before they take hold. It matches today’s pace: quick, responsive, built from real-time data instead of outdated assumptions},
        keywords = {Random Forest, XG Boost.},
        month = {April},
        }

Cite This Article

Samaleti, V., & Kumar, Y. A., & Nikhil, S. S., & Reddy, D. L. (2026). ML – Driven Real Time Classification And Prediction Of Ddos Attacks. International Journal of Innovative Research in Technology (IJIRT), 12(11), 1399–1403.

Related Articles