Towards Eco-Resilient Networks: CNN-Enabled Anomaly Detection for Green Cybersecurity

Cybersecurity is a major concern across various industries today. It is directly related to global goals such as enhancing infrastructure and industrial innovation (SDG 9) and fostering peace, justice, and strong institutions (SDG 16). Traditional problem-detection methods are becoming less effective as cyber threats evolve and increase in frequency, necessitating new approaches. This article introduces a novel technique for detecting abnormal patterns in network traffic using Convolutional Neural Networks (CNNs). The model is trained and tested on the CICIDS2017 dataset, which includes a wide range of both normal and suspicious network activities. Our CNN-based approach achieved a remarkable accuracy of 93.64%, outperforming conventional techniques. This method not only enhances cybersecurity but also contributes to SDG 9 by promoting resilient infrastructure. Additionally, it aids in building robust institutions by swiftly detecting and mitigating cyber threats, aligning with SDG 16. The performance of the model, including metrics like accuracy, recall, F1 score, and confusion matrices, has been thoroughly analyzed. These results underscore the importance of leveraging advanced machine learning techniques, especially deep learning, in cybersecurity solutions. Creating secure and sustainable digital environments depends heavily on integrating such innovations. Our model’s ability to accurately detect network anomalies demonstrates the potential of CNNs to revolutionize cybersecurity practices. This study advocates for the continued application of deep learning methods to address the growing challenges in cybersecurity, contributing to the achievement of key sustainable development goals while ensuring the stability and security of digital systems.