🛡️ AI-Based Intrusion Detection System

An AI-powered Intrusion Detection System (IDS) built using Python, Streamlit, and Scikit-learn that analyzes network traffic logs and classifies sessions as Normal or Suspicious.

---

Features

• Upload network log CSV files
• Machine learning-based intrusion detection
• Displays attack probability for every session
• Classifies traffic into:
  • ✅ Normal
  • ⚠️ Suspicious
• Interactive web interface using Streamlit

---

Technologies Used

• Python
• Streamlit
• Pandas
• Scikit-learn
• Joblib

---

Project Structure

Intrusion-detection/
│
├── app.py
├── requirements.txt
├── intrusion_pipeline.pkl
├── sample_input.csv
├── README.md
└── screenshots/

---

Installation

Clone the repository:

git clone https://github.com/AsjadAbbas995/Intrusion-detection.git

Go into the project directory:

cd Intrusion-detection

Install dependencies:

pip install -r requirements.txt

---

Running the Application

Start the Streamlit application:

streamlit run app.py

The application will open in your browser.

---

Input Format

The uploaded CSV should contain the same feature columns used during model training.

Example:

session_id	feature_1	feature_2	...

---

Output

The application generates:

• Attack Probability
• Prediction

Example:

Attack Probability	Prediction
0.95	Suspicious
0.12	Normal

---

Machine Learning Model

The application uses a pre-trained Scikit-learn pipeline stored as:

intrusion_pipeline.pkl

The model predicts the probability that a network session is malicious.

---

Requirements

• Python 3.10+
• Streamlit
• Pandas
• Scikit-learn
• Joblib

Install all dependencies using:

pip install -r requirements.txt

---

Future Improvements

• Real-time packet monitoring
• Multiple ML model comparison
• Threat severity scoring
• Visualization dashboard
• Explainable AI (SHAP/LIME)
• Database logging
• REST API integration

---

Author

Asjad Abbas

Cybersecurity Student | AI & Machine Learning | Penetration Testing