Authors - Rohan Anantapur, Pranav Rao Pernankil, Animesh Giri Abstract - The emergence of smart carts in the retail and shopping industry has introduced questions regarding the safety and security of these autonomous guided vehicles. While smart carts enhance the shopping experience through automation and real-time transaction processing, they are susceptible to various risks that can compromise their functionality and user data. Many existing cyberattack detection systems fail to consider a well-rounded set of features, leaving smart cart networks exposed to sophisticated threats. This approach enhances the resilience of smart cart systems by detecting and mitigating security breaches effectively. In this paper, we introduce DeepCart, a cyberattack detection and classification system that focuses on violations in smart cart networks. DeepCart leverages Generative Adversarial Networks (GAN) to identify cyberattack data patterns and employs Deep Belief Networks (DBN) for accurate classification of detected threats.
