Attacks Detection Model Based on a Machine Learning Algorithm

With the rapid developments in data volume and the complexity of cyber-attacks, attack detection systems face increasing challenges. This paper presents a model based on a Support Vector Machine (SVM) algorithm to detect attacks. To improve detection accuracy and reduce computational complexity, the Principal Component Analysis (PCA) algorithm was used as a first stage to select the most important features in the NSL-KDD dataset. The proposed model was applied to the NSL-KDD dataset. The results showed that using the SVM and PCA algorithms helps reduce the data dimensions, leading to improved classification accuracy and reduced computational complexity of the model. This model provides a machine learning-based detection system that can effectively identify attacks, which leads to enhancing network security against complex threats.