ATTENTION-AUGMENTED DEEP CONVOLUTIONAL FRAMEWORK FOR INTELLIGENT NETWORK TRAFFIC ANOMALY DETECTION

Abstract

The exponential growth of internet-connected devices has led to increasingly complex and high-volume network traffic, making anomaly detection a crucial challenge in cybersecurity. This research proposes an attention-augmented deep convolutional framework for intelligent network traffic anomaly detection. The model integrates attention mechanisms with big-step convolutional neural networks (CNNs) to enhance feature extraction and focus on critical traffic attributes. By leveraging datasets such as CICIDS2017, the system learns to differentiate between normal and malicious patterns efficiently. Experimental results show that the proposed model significantly improves detection accuracy, precision, and recall compared to traditional machine learning and basic CNN-based approaches. This framework demonstrates scalability, robustness, and adaptability, offering a promising solution for next-generation intrusion detection systems in modern digital networks