🛡️ Network Intrusion Detection System (NIDS)

📌 Overview

A Machine Learning–based Network Intrusion Detection System (NIDS) that classifies network traffic as normal or malicious.
The system uses classical machine learning and deep learning models trained on the NSL-KDD benchmark dataset to detect network attacks.

Designed with a modular pipeline focusing on preprocessing, model training, evaluation, and comparison.

---

🚀 Features

• Binary classification: Normal vs Attack
• Classical and deep learning models
• Trained on NSL-KDD dataset
• Model comparison and analysis
• Modular and extensible ML pipeline

---

🧠 Models Used

• LSTM (Long Short-Term Memory)
• RNN (Recurrent Neural Network)
• KNN (K-Nearest Neighbors) — baseline model

Deep learning models capture sequential patterns in network traffic, improving detection over traditional ML approaches.

---

📊 Dataset

• Name: NSL-KDD
• Type: Intrusion detection benchmark dataset
• Features: 41 network traffic attributes
• Classes: Normal traffic and multiple attack categories

---

🛠️ Tech Stack

• Language: Python
• Libraries: NumPy, Pandas, Scikit-learn
• Deep Learning: TensorFlow / Keras
• Tools: Jupyter Notebook, Google Colab

---

⚙️ Pipeline

1. Data preprocessing and feature encoding
2. Train-test split
3. Model training (KNN, RNN, LSTM)
4. Evaluation and comparison

---

📈 Results

Model	Accuracy
KNN	XX%
RNN	XX%
LSTM	XX%

Exact evaluation metrics can be updated after final experimentation.

---

📂 Project Structure

NIDS/ ├── data/ ├── preprocessing/ ├── models/ ├── evaluation/ ├── notebooks/ └── README.md

---

🔮 Future Work

• Real-time packet capture and classification
• Live network traffic integration
• Low-latency model optimization
• API-based deployment

---

🎯 Use Cases

• Enterprise network security
• Intrusion detection research
• Foundation for real-time IDS deployment

---

👤 Author

Het Bhutak
AI/ML Engineer

• GitHub: https://github.com/Hettbhutak
• LinkedIn: https://linkedin.com/in/het-bhutak-3101601ba