Heuristic Optimization Algorithm with Ensemble Learning Model for Intelligent Intrusion Detection and Classification

Abstract

Intrusion Detection (ID) for network security prevents and detects malicious behaviours or unauthorized activities that occurs in the network. An ID System (IDS) refers to a safety tool that monitors events or network traffic for responding to and identifying illegal access attempts or malevolent activities. IDS had a vital role in network security by finding and alerting security teams or administrators about security breaches or potential intrusions. Machine Learning (ML) methods are utilized for ID by training methods for recognizing behaviours and patterns linked with intrusions. Deep Learning (DL) methods are implemented to learn complicated representations and patterns in network data. DL methods have witnessed promising outcomes in identifying network intrusions by automatically learning discriminatory features from raw network traffic. This article presents a new Teaching and Learning based Optimization with Ensemble Learning Model for Intelligent Intrusion Detection and Classification (TLBOEL-IDC) technique. The presented TLBOEL-IDC method mainly detects and classifies the intrusions in the network. To attain this, the TLBOEL-IDC method primarily preprocesses the input networking data. Besides, the TLBOEL-IDC technique involves the design of an ensemble classifier by the integration of three DL models called Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Bidirectional LSTM (BLSTM). Moreover, the hyperparameter tuning of the DL models takes place using the TLBO approach that improves the overall ID outputs. The simulation assessment of the TLBOEL-IDC approach takes place on a benchmark dataset and the outputs are measured under various factors. The comparative evaluation emphasized the best accomplishment of the TLBOEL-IDC technique over other present models by means of diverse metrics.