Project Name: DDoS Attack Detection in SDN Using Multiple Machine Learning Models

Project Overview: This project aims to enhance the detection of Distributed Denial of Service (DDoS) attacks in Software Defined Networking (SDN) environments by leveraging machine learning techniques. The study compares the performance of different models to determine the most effective approach for accurate and efficient DDoS detection.

Dataset: The dataset used for this project is the CIC-DDoS2019 dataset. The dataset has been preprocessed and cleaned to remove any inconsistencies and ensure it is suitable for training the models.

Models Implemented:

• Support Vector Machine (SVM): Evaluated with linear, polynomial, and RBF kernels.
• K-Nearest Neighbors (KNN): Tested with Euclidean, Manhattan, and Cosine distance metrics.
• Naive Bayes (NB): Gaussian Naive Bayes.
• Ensemble Model: Combines SVM, KNN, and NB using soft voting.

Feature Selection: Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) were used for feature selection to improve model accuracy and computational efficiency. PCA was used for SVM, KNN, and the Ensemble model, while LDA was used for Naive Bayes. The feature selection process helped reduce the dimensionality of the dataset while retaining the most important features.

Hyperparameter Tuning: Halving Grid Search was used for hyperparameter tuning to find the best parameters for each model, optimizing their performance.

Results: The results showed that the Ensemble model achieved the highest accuracy of 98.29%, outperforming individual models. The performance metrics for each model are as follows:

• SVM (RBF kernel): 94.67% accuracy
• KNN (Manhattan distance): 93.37% accuracy
• Naive Bayes: 80.19% accuracy
• Ensemble Model: 98.29% accuracy

View on GitHub