Project 04 / Machine Learning / Cybersecurity / Research

DDoSDetection

2025Python/Deep Learning/Multitask Learning

RESEARCHAcademic thesis work, presented at an international conference rather than shipped as a product.

System path

NETWORK TRAFFIC
  |
  v
MULTITASK LEARNING
  |
  v
DETECTION

Overview

"Distributed Denial-of-Service (DDoS) Detection Using Multitask Learning and Deep Learning" was my undergraduate thesis at Telkom University, exploring how a multitask deep learning approach can be used to detect DDoS attacks. The work was presented at ICAIIC 2025, an international AI and communications conference held in Japan.

My Role

This was an independent research project. I framed the research problem, implemented the detection approach in Python, and ran the experimentation behind the thesis, under academic supervision at Telkom University.

Technology

Python/Deep Learning/Multitask Learning

Engineering

The project applies a multitask learning framework to DDoS detection: rather than training a single model for one classification task, a multitask approach lets the model learn related detection objectives together. Specific architecture and dataset details are documented in the full thesis and conference paper rather than reproduced here, to avoid misrepresenting the technical specifics outside their original context.

Outcome

The work was accepted and presented at ICAIIC 2025 (International Conference on Artificial Intelligence in Information and Communication), held in Japan.

Reflection

This project pushed me from applying machine learning to actually researching it: framing a problem, defending a methodology, and presenting it to an academic audience. It's also where my interest in cybersecurity and applied ML started to take shape as a real direction, not just a course requirement.

Back to selected work