<![CDATA[The increasing intensity and complexity of cyber threats demand more adaptive intrusion detection mechanisms. Conventional approaches are often limited in capturing complex and non-linear attack patterns in network traffic data. This study develops and evaluates a convolutional neural network (CNN)-based model for multi-class cyberattack detection. The proposed architecture integrates convolutional, pooling, and fully connected layers with rectified linear unit (ReLU) and SoftMax activation functions to improve classification performance. The network security laboratory-knowledge discovery and data mining (NSL-KDD) dataset is used for training and evaluation. Experimental results show that the CNN model achieves 96.34% accuracy and an F1-score of 0.99, outperforming several traditional machine learning methods, including Naïve Bayes (NB), decision tree (DT), support vector machine (SVM), and random forest (RF). The superior performance is attributed to the model’s capability to automatically learn and extract meaningful spatial representations from network data without manual feature engineering. These findings demonstrate the effectiveness of deep learning techniques in improving cyberattack detection and contribute to the development of reliable AI-driven network security systems with strong potential for real-world cybersecurity applications and evolving threat mitigation strategies.]]>
Copyrights © 2026
