Privacy-Preserving Deep Learning Techniques for Business Big Data
M Spoorthi, Priyanka Mohan, Harinahalli Lokesh Gururaj, Jaroslav Frnda · 2025
As enterprises increasingly utilize big data for analytics and decision-making, safeguarding data privacy has emerged as a critical concern. This chapter examines the convergence of machine learning (ML) and deep learning (DL) with privacy-preserving methodologies within the realm of business big data. The analysis commences by exploring the distinctive privacy concerns presented by commercial datasets, frequently include sensitive consumer and corporate information. We examine several privacy-preserving techniques, such as differential privacy, federated learning, homomorphic encryption, and secure multiparty computation. This chapter analyzes different strategies, evaluating the trade-offs among data utility, privacy, scalability, and security. Implementation frameworks, tools, and best practices are provided to assist firms in adopting privacy-preserving analytics. The chapter concludes by discussing future developments, such as the amalgamation of privacy-preserving techniques with edge computing and the rise of Machine-Learning-as-a-Service (MLaaS) offerings. This chapter offers a thorough examination of privacy-preserving ML and DL techniques, intending to furnish academics, practitioners, and policymakers with the requisite expertise to protect sensitive corporate data in the age of big data analytics.