A Python + Flask web application that detects anomalies in logs using machine learning (Isolation Forest) and visualizes them for easy debugging.

• Upload log files (.log or .txt)
• Intelligent anomaly detection beyond simple ERROR filtering
• Advanced feature extraction from log messages
• Visualization with highlighted anomalies
• Export results as CSV for further analysis

• Python 3.10+
• Flask for the web interface
• scikit-learn for Isolation Forest algorithm
• pandas for data manipulation
• matplotlib for visualization

1. Clone the repository

   git clone https://github.com/Harivelu0/log-anomaly-detector.git
   cd log-anomaly-detector

2. Create & activate virtual environment

   python3 -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

   pip install -r requirements.txt

4. Run the application

   python main.py

5. Open http://127.0.0.1:5000/ in your browser

log_analyser/
├── Readme.md
├── core/
│   ├── anamoly_detector.py  # ML algorithm implementation
│   ├── parser.py            # Log file parsing
│   └── preprocessor.py      # Feature extraction
├── logs/                    # Uploaded logs storage
├── main.py                  # Flask application
├── requirements.txt         # Dependencies
├── static/                  # Static assets (charts)
│   └── anomaly_chart.png
└── templates/               # HTML templates
    └── index.html           # Main UI

1. Log Parsing: Convert raw logs to structured data
2. Feature Extraction: Generate numeric features from log content
3. Anomaly Detection: Use Isolation Forest to identify outliers
4. Visualization: Display anomalies in chart and table format

1. Open the web UI
2. Upload a log file (.log or .txt)
3. The app will:
   • Parse the logs
   • Preprocess log messages
   • Detect anomalies using ML
   • Show results in a chart and a table
4. You can download the results as anomalies.csv

• Add support for more log formats
• Implement adaptive learning based on user feedback
• Add clustering to group similar anomalies
• Create API endpoints for integration with other systems
• Add real-time processing capability for streaming logs

Made with ❤️ by Hari