Cybersecurity Using Hybrid Type Model for Classification Through SCO Optimization Technique

The effectiveness of Network Intrusion Detection Systems (NIDS) in recognizing complex attacks is hindered by evasion strategies like polymorphic malware and encrypted traffic. Vulnerabilities in NIDS algorithms are regularly exploited by cybercriminals, highlighting the dynamic relationship between changing attack tactics and cybersecurity defenses. An enhanced approach to strengthen Network Intrusion Detection Systems is suggested by this study. The methodology begins with a rigorous preprocessing phase that includes normalization functions to progress the accuracy and consistency of the input data. Optimizing the NIDS's efficacy is the aim of the Single Candidate Optimization (SCO) algorithm for feature selection. We utilize a hybrid model that incorporates ANN, Wavelet Transform, and Long Short-Term Memory (LSTM) for the classification phase. This model is designed to identify sequential dependencies in the network traffic data that we have. A second iteration of the SCO is devoted to hyperparameter optimization in order to attain optimal model performance and to further optimize the model. For our review, we opted to use the BoT-IoT dataset because it is a gold standard in the field. The suggested approach demonstrates how SCO can improve hyperparameter optimization and feature selection, leading to a more accurate and cyberattack-resistant NIDS. Consequences from experiments and evaluations of performance demonstrate that the proposed strategy is effective; it achieves a remarkable 99.6 percent accuracy. This proves how successful it is in judgement to current representations, which is a significant step towards consolidation cybersecurity defenses against changing threat landscapes.