Classification of Inflammatory Gene Expression Patterns with Machine Learning Models
Image generated by DALL-E
Machine Learning (ML)-based Network Intrusion Detection Systems (NIDSs) have proven to become a crucial technique for detecting malicious network activities from cyber criminals. This research leverages a decision tree to differentiate and classify between normal network activities and invasions. The model is trained with the UNSW-NB15 dataset. While each data sample comprises many features, not all are discriminative in the classification task. An ensemble multi-objective Harris’ hawk optimiser is designed and developed to optimise the model with multiple objectives, viz., minimising the number of features, maximising sensitivity, and maximising specificity.