Machine Learning Approach for Network Cyber Intrusion Detection

Nowadays, everyone is interconnected through the Internet for exchanging digital information. This information is stored using cloud technology. However, the rapid growth of cloud technology has led to an accumulation of the volume of digital data, as well as network intrusions. Consequently, protecting this data has become crucial for various reasons. Therefore, this study presented a method for detecting network cyber intrusions. The instances of network cyber intrusions were gathered from the KDD’99 Cup database. Furthermore, the proposed method employed the Gaussian Naïve Bayes (GNB) approach to identify instances of cyberattacks. The proposed method utilized various measurements for the purpose of generally assessing the performance of the GNB classifier. The experimental results have been demonstrated that the proposed GNB classifier has achieved 94.28% accuracy in the detection of network attacks. In addition, the GNB has achieved 98.32% precision, 94.28% sensitivity, and 95.89% F-measure. The proposed GNB algorithm demonstrated its efficiency in detecting network attacks, outperforming its counterparts in terms of detection accuracy.