|
An Intelligent IoT based anomaly detection using predictive analytics |
|---|---|
| รหัสดีโอไอ | |
| Creator | Vergin Sarobin |
| Title | An Intelligent IoT based anomaly detection using predictive analytics |
| Contributor | Ranjith Jayaramu, Shree Varshan, Briju Demel, Hridhay Vardhan |
| Publisher | Khon Kaen University, Thailand |
| Publication Year | 2568 |
| Journal Title | Asia-Pacific Journal of Science and Technology |
| Journal Vol. | 30 |
| Journal No. | 6 |
| Page no. | 12 (14 pages) |
| Keyword | Anomaly detection, Classification model, IoT-23, Machine learning, Security |
| URL Website | https://apst.kku.ac.th/ |
| Website title | https://apst.kku.ac.th/an-intelligent-iot-based-anomaly-detection-using-predictive-analytics/ |
| ISSN | 2539-6293 |
| Abstract | In this study, we analyze the effectiveness of combining various machine learning approaches in order to detect specific attack classes within the Internet of Things (IoT). Using the IoT23 dataset, the research examines distinct features associated with attack classes derived from the raw data. Multiple algorithms, such as Multi-layer Perceptron, Random Forest, Extreme Gradient Boosting (XGBoost), Decision Tree, K-Nearest Neighbor, Support Vector Machine, Logistic Regression, and Naive Bayes, were thoroughly trained and evaluated. Through the implementation of ensemble learning techniques, the study successfully achieved an elevated detection rate of attack classes and an overall accuracy improvement, maintaining a false alarm rate of up to 15%. The research highlights the importance of using ensemble learning methods to identify and categorize attacks in IoT domains, serving as a valuable resource for further research. The insights revealed in this study provide readers with a compelling reason to read it and should act as a catalyst for further research in similar directions. |
