Machine Learning Based Anomaly Detection for DDoS Protection
The project's core focus is on developing innovative anomaly detection algorithms and machine learning models specifically tailored for DDoS (Distributed Denial of Service) protection.
Key Responsibilities:
· Engage collaboratively to conceptualize, design, and implement state-of-the-art machine learning models aimed at detecting anomalies in web requests.
· Utilize and analyze a benchmark dataset comprising of website visitor sessions,
along with creating additional attack simulations to enhance model robustness.
· Develop, maintain, and refine Python scripts for various tasks including data
preprocessing, feature generation, model training, and algorithm optimization.
· Perform thorough performance evaluations of the developed models and
algorithms, ensuring both accuracy and efficiency are upheld.
· Document the project's progress and findings diligently, providing insightful reports
and research outcomes.
Machine Learning Models to be Utilized: Isolation Forest, Decision Trees, Support Vector Machines, Hidden Markov Models, Feed-Forward Neural Networks, Long Short-Term Memory Networks (LSTMs), Gated Recurrent Units (GRUs)
Required knowledge
- Current enrollment or recent graduation from a program in Computer Science,
Engineering, Data Science, or a related field.
· Strong proficiency in Python programming, with hands-on experience in libraries
like Scikit-Learn, TensorFlow, Keras, or PyTorch.
· A foundational understanding of machine learning concepts and algorithms.
· Familiarity with network security and DDoS attacks will be considered a valuable
asset.