Random Forest Hyperparameter Tuning in Machine Learning for Improved Performance in Intrusion Detection Systems

Abstract

In recent times, researchers have been proposing different Intrusion Detection methods to deal with the increasing number and complexity of threats as technology keeps emerging. In this context, Random Forest models have been providing a notable performance on her predictive capacity to applications in the realm of behavioural-based Intrusion Detection Systems and other related fields of specialization which includes medicines, Banking, commerce, etc in terms high magnitude forecasting and optimal predictions . In this work, in-depth evaluation analysis of the Random Forest tuning were carried out with respect to classification, feature selection, and proximity metrics. This empirical research will provide an inclusive review of the general basic concepts related to Intrusion Detection Systems, which includes taxonomies, data collection, modeling and evaluation metrics. Furthermore, the manual hyperparameter tuning technique was used for this research work and a desirable experimental output was achieved as showed in this work.