Network Intrusion Detection Models based on Naives Bayes and C4.5 Algorithms
Olamatanmi Josephine Mebawondu, Olugbemiga Solomon Popoọla, Ikechukwu Ignatius Ayogu, Chukwuemeka Christian Ugwu, Adebayọ Olusọla Adetunmbi
Abstract
All around the world, the rapid spread of the pandemic (COVID-19) has brought an enormous challenge, especially to the ICT industry. The total lockdown which prevailed had increased the use of the internet, which is a challenge to safety and security. Thus, an Intrusion Detection System (IDS) is needed to maintain this emergence of the boundless communication paradigm. This paper proposed an optimized Network IDS by applying two machine learning algorithms in intrusion dataset and feature selection techniques to optimize the IDS model. The viability of this work is shown by comparing, the result of the model with existing work. The decision tree applied outperformed the Naïve Bayes algorithm with 89.27% and 75.09% accuracy, respectively.