Advancements in Machine Learning Techniques for Intrusion Detection Systems: An Overview of Perspectives and Datasets

This paper offers a comprehensive study on the topic of intrusion detection systems (IDS) in the context of cyber security, focusing on the application of machine learning (ML). A range of ML methods are explored, including logistic regression (LR), Bayesian logic, support vector machine (SVM), and convolutional neural network (CNN), among others. The study considers various datasets used in IDS and evaluates the advantages and disadvantages of each model. The paper also discusses new approaches that have emerged since 2020. To assess the accuracy of the models, the study compares their performance in supervised and unsupervised classification tasks and ranks them based on key metrics such as detection rate, false alarm, and accuracy. The study identifies the most effective algorithm for IDS in cyber security and explains the rationale behind this choice. Overall, this study provides valuable insights into the application of ML for intrusion detection in cyber security and serves as a practical guide for researchers and practitioners in the field.