Advanced Intrusion Detection System Leveraging Support Vector Machine algorithm

Aim: The aim of this study is to evaluate the effectiveness of a K-Nearest Neighbours (KNN) technique and an algorithm based on Support Vector Machines (SVM) for improving intrusion detection systems' accuracy. Materials and Methods: Group 1 used a model based on KNN. Using a simulated network environment, metrics like scalability, processing time, and accuracy were evaluated. Group 2 classified network traffic using an SVM-based model. At a 95% confidence interval, with a significance level of 0.05% and a G Power value of 80%. Result: In terms of detection accuracy, the SVM-based model in Group 1 achieved 96%, outperforming the KNN-based model in Group 2, which had an accuracy of 88%. Conclusion: The SVM-based algorithm outperformed the KNN model in terms of accuracy,speed,and scalability,making it a superior choice for intrusion detection systems.Its deployment can enhance network security by improving anomaly detection rates and reducing false positives,thereby supporting effective threat mitigation strategies.