Autor: |
Merlin, R. Tino, Ravi, R. |
Předmět: |
|
Zdroj: |
Journal of Namibian Studies; 2023 Special Issue, Vol. 38, p2015-2034, 20p |
Abstrakt: |
As IoT devices become increasingly common in smart cities, securing these interconnected networks has become essential. This study proposes an extensive technique for improving intrusion detection systems designed especially for IoT networks in smart city contexts. Leveraging machine learning techniques and preprocessing methodologies, the study addresses the challenge of detecting and mitigating potential security threats. The methodology encompasses various preprocessing steps, including Synthetic Minority Oversampling Technique (SMOTE) employed for class imbalances and Min-Max normalization for scaling feature values. Additionally, advanced feature selection techniques such as Recursive Feature Elimination with Cross-Validation (RFECV) and the Boruta algorithm are employed to identify the most relevant subset of features for accurate intrusion detection. The accuracy achieved by the proposed methodology is 97.62%. [ABSTRACT FROM AUTHOR] |
Databáze: |
Complementary Index |
Externí odkaz: |
|