🔍 Advanced Anomaly Detection System (AADS)

An end-to-end Machine Learning pipeline designed to monitor real-time data streams and flag statistical irregularities. This system integrates a predictive model with a REST API and a simulation dashboard to visualize anomaly detection as it happens.

---

🏗️ System Architecture

The project is divided into four core components:

• Model Trainer (train_model.py): Script to preprocess data and train the detection algorithm.
• Inference Engine (anomaly_api.py): A Flask/FastAPI backend that serves the trained model via REST endpoints.
• Real-time Simulator (realtime_simulator.py): A script that generates synthetic data streams to test the model's limits.
• Web Dashboard: Located in templates and statics, providing a visual representation of the data flow and anomaly spikes.

---

📈 Real-Time Monitoring Output

The system evaluates incoming values and assigns a classification based on the learned decision boundary.

Sample Detection Logs:

Timestamp	Data Value	Status	Label
21:55:06	-0.7259	✅ Normal	model=0
21:55:04	0.0103	✅ Normal	model=0
21:55:03	-5.3221	🚨 Anomaly	model=1
21:55:02	5.4508	🚨 Anomaly	model=1

---

🚀 Quick Start

1. Environment Setup

Clone the repository and initialize a virtual environment:

git clone https://github.com/Vbhhacl/advanced-anomaly-detection-system.git
cd advanced-anomaly-detection-system
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

2. Install Dependencies

pip install -r requirements.txt

3. Execution

First, ensure the model is trained, then start the API and Simulator:

python train_model.py
python anomaly_api.py
python realtime_simulator.py

---

📂 Project Structure

├── statics/             # CSS and JS for the dashboard
├── templates/           # HTML templates
├── anomaly_api.py       # API for model serving
├── anomaly_model.pkl    # Serialized ML model
├── realtime_simulator.py # Data stream generator
├── requirements.txt     # Project dependencies
├── train_model.py       # Model training logic
└── .gitignore           # Version control exclusions

🛠️ Tech Stack

• Language: Python 3.x
• ML Libraries: Scikit-Learn, NumPy, Pandas
• API Framework: Flask / FastAPI
• Frontend: HTML5, CSS3, JavaScript

---

Author: Vbhhacl