Development of Deep Learning Models for Analysis of Network Traffic to Detect and Classify an Cyber-attack

Abstract

Security in network infrastructures, especially in current times, takes top priority. Techniques to detect and mitigate malicious activities pose a serious challenge. This research work is defined as developing deep learning models to analyze network traffic data for detecting and classifying abnormal behavior that could lead to cybersecurity threats. Using artificial neural networks, deep learning algorithms help to find anomalies such as DoS attacks, malware presence, and data exfiltration in real-time. The proposed models have been trained with different datasets of various attack vectors to detect anomalies with high precision. The research work demonstrates that deep learning models effectively differentiate between normal and abnormal network traffic. Hence, it provides a foundation for real-time network security monitoring and threat mitigation. The study signifies the role of AI in securing networks against cyber-attacks and opens doors for future research in the development of automated threat detection systems.