Integrated Computer Network Security System: Intrusion Detection and Threat Prediction Using Machine Learning Algorithms

The escalating use of computer networks and associated applications has made cybersecurity a critical concern. This research presents a comprehensive approach to addressing network security issues by developing an Intrusion Detection System (IDS) based on the dataset from the Security Information and Event Management System (SIEM) using various machine learning algorithms. The study includes a detailed comparison of models such as Logistic Regression, K-Nearest Neighbors, Gusain Naive Bayes, Support Vector Machine, Decision Tree, Random Forest, XGBoost, and Artificial Neural Networks. The results demonstrated using histograms and tables show the effectiveness of Random Forest and PCA Random Forest, emphasizing their accurate traffic classification. The research evaluates the model’s performance across various cyber-security datasets containing multiple categories of cyber-attacks. It evaluates efficiency by utilizing criteria such as accuracy, precision, and recall. By applying a multilevel approach aligned with the latest trends in machine learning, the study aims to facilitate swift and precise threat analysis and response, ultimately enhancing the overall effectiveness of the cybersecurity system. The study is an effective educational resource that introduces IT students to innovative machine learning and cybersecurity ideas. Through the integration of the results into IT curriculum, educators may close the knowledge gap between theory and practical application by offering students real-world experience in addressing cyber threats. This combined emphasis on innovation and education seeks to develop a new generation of IT professionals with the expertise required for enhanced cybersecurity.