To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
A Machine Learning Framework for Automatic Detection and Classification of Cyber Attacks in IoT Use Cases |
|
Author |
Hussam Aleem Mohammed Amar Yusof Jaffar |
| Citation |
Vol. 25 No. 1 pp. 148-158 |
|
Abstract |
Internet of Things (IoT) use cases are vulnerable to cyber-attacks due to lack of global standards and involvement of heterogeneous devices, protocols and platforms. Traditional methods are found inadequate safeguard IoT applications. With the emergence of Artificial Intelligence (AI), machine learning (ML) and deep learning techniques are widely used to solve security problems in different applications. Learning capability of AI models paves way for intelligent solutions. In this paper, we proposed a ML framework for automatic detection and classification of cyber-attacks in IoT use cases. We proposed a hyperparameter optimization method, designed for optimization of parameters of four ML techniques in tune with the dataset, used in the proposed framework. An algorithm named Learning based Optimal Machine Learning for Cyber Attack Detection and Classification (LbOML-CADC) is also proposed. This algorithm exploits hyperparameter tuning method for efficient detection and classification of cyber-attacks. We evaluated our framework using UNSW-NB15 dataset. Our empirical study reveals that highest accuracy achieved is 97.59%. |
|
Keywords |
Security, Machine Learning, Cyber Attack Detection, Hyperparameter Optimization, Internet of Things |
|
URL |
