A Effective Detection Method using Optimized Decision Tree Classification for Network based Intrusion

In today's world, the rapid evolution of the internet makes things very difficult. Network security also faces the challenge of protecting computer networks, one of the most challenging tasks in the field. Communications have often been attacked by intruders. To create a solution for detection of intrusions, research is thus essential. To detect these intrusions, there have been many solutions created. In terms of creating the solution for detecting abnormal activities on a network, Intrusion Detection Systems (IDS) are one of the most important. There are also numerous challenges to identify the intrusion for accurate detection and false alarm rates for detection, so that it is not completely given the solution. Hence, the research is essential to improve the IDS technique in order to provide accurate solutions. Machine Learning algorithms have been developed in a way that solves IDS-based problems effectively. Accordingly, the research work has chosen Network Based Intrusion Detection System (NIDS); it defeats intrusion issues with the aid of improved preprocessing and Optimized Decision Tree Classifier (ODT). Python was used in the implementation of the research. The following performance metrics will be utilized in this research work in order to prove an experimented result: Precision, Recall, F1-Score and Accuracy. Comparatively, the overall work can be compared to existing machine learning algorithms, such as Naïve Bayes (NB), Logistic Regression (LR), Linear Discriminant Analysis (LDA), and K-Nearest Neighbor (KNN). At the end, this research work achieved the greatest accuracy out of all of them.