Explainable AI, Federated Learning, and AI Ethics in E-Commerce
Gunjan Garg, Chander Prabha, Bhavisha Ahuja · Advances in electronic commerce (AEC) book series/Advances in electronic commerce series · 2024
E-commerce data security is influenced by the three elements; Explainable AI (XAI), Federated Learning, and Artificial Intelligence (AI) Ethics. The objective of Explainable AI is to enhance the transparency of AI systems and their explainability to promote human oversight and foster confidence. Federated Learning (FL) provides a decentralized approach by training Machine Learning (ML) models to secure user data confidentiality and privacy. Instead of exchanging local data, model constraints from local ML models are pooled and shared as a method of collaborative learning among devices and organizations. The bias, fairness, and the effects on society are a few moral issues surrounding the use of AI covered in AI Ethics. It discusses moral regulations, guidelines, and policies that support the development and application of moral AI. This chapter aims to give insight into the technological and ethical aspects of AI advances concerning AI-ML for data security in E-Commerce, with a focus on the importance of privacy, accountability, and transparency in AI systems.