A Multi-Objective Genetic GAN Oversampling: Application to Intelligent Transport Anomaly Detection\

The Internet of Things (IoT) enables the automation of data collection and processing functions but exposes a huge amount of data to the cyberattacks risk. To tackle this issue, anomaly detection allows to identify data points, events, and/or observations that deviate from a dataset's normal behaviour indicating eventual critical incidents. In this paper, we focus on the imbalance data and the minority classes problem where the number of abnormal samples is much less than normal (secure) samples. In particular, this paper presents a new equilibrium model based on a Genetic Algorithm to improve Generative Adversarial networks (GANs). This model addresses the problem of class imbalance to anomaly detection system performance. The proposed approach use is illustrated by a case study: an intelligent transport system-based scenario.