https://github.com/onome-joseph/anomaly-detection
Comprehensive anomaly detection algorithm designed to analyze and identify anomalies in computer internal features.
https://github.com/onome-joseph/anomaly-detection
anomaly-detection-algorithm hardware-monitoring python z-score
Last synced: 27 days ago
JSON representation
Comprehensive anomaly detection algorithm designed to analyze and identify anomalies in computer internal features.
- Host: GitHub
- URL: https://github.com/onome-joseph/anomaly-detection
- Owner: Onome-Joseph
- License: mit
- Created: 2025-01-12T22:28:47.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2025-01-12T22:58:19.000Z (4 months ago)
- Last Synced: 2025-02-12T06:39:08.933Z (3 months ago)
- Topics: anomaly-detection-algorithm, hardware-monitoring, python, z-score
- Language: Jupyter Notebook
- Homepage:
- Size: 294 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
## Anomaly Detection Algorithm for Computer Internal Features
This repository contains a comprehensive anomaly detection algorithm designed to analyze and identify anomalies in computer internal features such as:
- CPU usage
- Temperature
- Battery status
- Memory usage
---
The dataset comprises **1GB of synthetic multivariate time series data**, collected from the internal features of a computer. It includes **8,712,000 samples**, intentionally augmented with extreme values to test and evaluate the performance of the anomaly detection algorithms.
## Algorithms Used
1. **Z-Score Method**:
- Detects anomalies by calculating the standard score of each data point.
- Data points with a z-score above or below a specified threshold are flagged as anomalies.
2. **Interquartile Range (IQR) Method**:
- Identifies anomalies by analyzing the interquartile range of the data.
- Data points falling outside the lower and upper bounds (calculated using the IQR) are considered anomalies.
### Visualization
The anomalies detected by the z-score method are visualized to provide clear insights into the data patterns and deviations. The visualizations highlight anomalies within the multivariate time series data for easier interpretation.
## Key Features
- Handles large-scale multivariate time series data effectively.
- Implements robust statistical methods for anomaly detection.
- Includes synthetic data with extreme values to test the reliability of the model.
## Applications
1. **System Monitoring**: Continuously monitor system performance and detect irregularities in real-time.
2. **Security**: Detect abnormal behavior that may indicate security breaches or potential threats.
3. **Performance Optimization**: Gain insights into resource usage patterns and optimize system performance.
## How to Use
1. Ensure OpenHardwareMonitor is Installed and Running:
```bash
https://openhardwaremonitor.org/
```
2. Clone this repository:
```bash
git clone https://github.com/Onnome-Joseph/Anomaly-detection.git
```
3. Run the anomaly detection script:
```bash
python src/anomaly_detection.ipynb
```
4. View the visualizations in the `visualizations/` folder.
## Future Work
- Extend the algorithm to include machine learning-based anomaly detection techniques.
- Integrate real-time data streaming and detection.
- Explore additional visualization techniques for better data analysis.
## Contributing
Contributions are welcome! Feel free to open an issue or submit a pull request for any improvements or suggestions.