Using Generative Adversarial Networks for Anomaly Detection in Network Traffic: Advancements in AI Cybersecurity
Ravindra Changala, S. Kayalvili, Mansoor Farooq, L Malleswara Rao, Vuda Sreenivasa Rao, S Muthuperumal
Abstract
Anomaly detection in network traffic is a critical aspect of cybersecurity, aiming to identify abnormal behaviors indicative of potential security threats. In recent years, advancements in AI techniques have led to the exploration of innovative approaches for enhancing anomaly detection capabilities. This study introduces a novel methodology leveraging Generative Adversarial Networks (GANs) for anomaly detection in network traffic data, contributing to the advancement of AI cybersecurity. The proposed approach integrates Principal Component Analysis (PCA) for feature selection, allowing for the extraction of essential network traffic patterns while reducing dimensionality. Subsequently, GANs are employed to learn the underlying distribution of normal network traffic and generate synthetic samples for anomaly detection. This framework offers a holistic solution that combines the strengths of feature selection and generative modeling to effectively discern anomalous activities from normal network behaviors. The methodology is evaluated using a comprehensive dataset collected from Kaggle, comprising diverse network traffic records. Results demonstrate the superior performance of the PCA-GAN method compared to existing techniques, achieving high accuracy and robust detection of anomalies in network traffic. The integration of PCA for dimensionality reduction and GANs for generative modeling proves to be effective in enhancing cybersecurity measures, providing a promising solution for detecting and mitigating security threats in network infrastructures. With an accuracy of around 99.12%, the suggested method which uses Python software outperforms other existing techniques like CNN-GRU (Gated Recurrent Unit), Convolutional Neural Network (CNN)-LSTM, and Conv-LSTM by 1.49%. This research contributes to the ongoing efforts in leveraging AI technologies for bolstering cybersecurity defenses and safeguarding critical systems against malicious activities.