Data Privacy Security

This course equips you to reason about privacy risks in modern data science pipelines and to design privacy-preserving systems for real applications. You will first study practical and heuristic approaches to privacy, including de-identification, linkage attacks, k-anonymity-style release strategies, privacy-preserving machine learning, federated learning, and the regulatory context that shapes responsible data use. You will then learn techniques with formal guarantees, including differential privacy, secure multi-party computation, zero-knowledge proof, and fully homomorphic encryption, and examine the trade-offs between privacy, utility, and efficiency. By the end of the course, you will be able to choose, justify, and prototype suitable privacy-enhancing technologies for data analytics and machine learning tasks.