NSL-KDD ML Model for Network Security Course

• Apply at least one optimization algorithm for feature selection:
  • Genetic Algorithm (GA)
  • Ant Colony Optimization (ACO)
  • Particle Swarm Optimization (PSO)
• Compare effectiveness of selected techniques in enhancing model performance.

• Classical Machine Learning Models:
  • Efficient implementation of:
    • Logistic Regression
    • Support Vector Machine (SVM)
    • k-Nearest Neighbors (KNN)
    • Decision Tree
    • Random Forest
• Neural Network Models:
  • Implement a basic Neural Network model.
  • Advanced Deep Learning Models:
    • Recurrent Neural Network (RNN):
      • Long Short-Term Memory (LSTM) or
      • Gated Recurrent Unit (GRU) with sequential data transformation.
    • Convolutional Neural Network (CNN):
      • Transform dataset to a format compatible with CNN.
  • (Optional) Deep Reinforcement Learning (DRL):
    • Convert dataset into states and actions.
    • Create environment and train DRL model.

• Existing Metrics:
  • Accuracy
  • Balanced Accuracy
  • Precision
  • Recall
  • F1-Score
  • Matthews Correlation Coefficient (MCC)
  • Model complexity (e.g., Training time, dataset size, etc.)
  • Adaptability (detect or infer new types of attacks)
• Additional Metrics:
  • Introduce two new relevant evaluation metrics.
  • Assess usefulness in the application.

• Detailed Documentation:
  • Data preprocessing steps and dataset description.
  • Explanation of preprocessing algorithms.
  • Feature selection methods and comparative analysis.
  • General description of models used.
  • Transformation techniques for LSTM, GRU, CNN, and DRL models.
  • Interpretation of results and conclusions.
• Excel Sheet:
  • Summary of performance metrics (columns) for all models (rows).
• Notebooks:
  • Provide all notebooks used for implementation.
• Dataset:
  • Include the dataset used.

1. Code Implementation:
   • Jupyter notebooks.
2. Documentation:
   • Word document with comprehensive explanations.
3. Excel Sheet:
   • Organized model performance metrics.
4. Dataset:
   • Dataset used for the project.