IEEE Communications Surveys & Tutorials, a top-tier journal in the fields of communications and computer science, has published a review paper on federated analytics by Associate Professor Yifei Zhu and his research team at the University of Michigan-Shanghai Jiao Tong University Joint Institute (UM-SJTU JI, JI hereafter). The paper, titled “A Survey on Federated Analytics: Taxonomy, Enabling Techniques, Applications, and Open Issues”, provides a systematic review of the emerging field of federated analytics, shedding light on its core technologies, applications, and unresolved challenges.
Massive data generated by edge devices have become critical to the functioning of modern applications, providing the foundation for intelligent algorithms that enhance user experiences. However, with growing concerns around data privacy and the increasing number of data privacy regulations worldwide, traditional centralized data analysis models are facing significant challenges. Federated analytics offers a solution by enabling collaborative data analysis without centralizing sensitive data, thus meeting privacy requirements while continuing to leverage valuable edge data. Although federated analytics has been widely researched and applied in academia and industry, many existing studies lack a systematic review of the field.
To address the gap, the JI research team delves into various enabling technologies that drive federated analytics, exploring its applications across domains like statistical metrics computation, frequent pattern mining, and database tasks. These examples underscore the vital role federated analytics plays in privacy-sensitive, large-scale distributed environments such as the Internet of Things (IoT) and Big Data. The team led by Professor Zhu also highlights critical areas for future exploration, including expanding application scenarios, innovating algorithms, optimizing system performance, and enhancing cross-layer collaboration. These insights provide a clear roadmap for continued advancement in federated analytics, supporting the development of privacy-preserving distributed data processing technologies.
The published paper, with Professor Yifei Zhu as the corresponding author, is first-authored by JI doctoral student Zibo Wang. Other co-authors include JI doctoral student Haichao Ji, Professor Dan Wang from The Hong Kong Polytechnic University, and Professor Zhu Han from the University of Houston. The research was supported by the Ministry of Science and Technology’s Key R&D Youth Scientist Program.
IEEE Communications Surveys & Tutorials is the highest-impact journal in the IEEE communications field (with a 2024 impact factor of 34.4) and one of the most influential journals in the telecommunications sector. The journal is dedicated to publishing review articles on cutting-edge technologies in the computer and communications fields, widely recognized by academia and industry as an important reference for the latest advancements and future trends in the field.
Author Introduction
Zibo Wang, Bachelor of Engineering from SJTU (Class of 2020), is currently a fifth-year Ph.D. student in Information and Communication Engineering at JI. He is a recipient of the National Scholarship for Graduate Students and an outstanding graduate at the university level. After graduation, he will join the Shenzhen National Laboratory as an Assistant Researcher.
Yifei Zhu is an Associate Professor at JI. He obtained his Ph.D. in Computer Science from Simon Fraser University in Canada. His research interests include edge computing, multimedia networks and systems, and distributed machine learning systems. In recent years, Professor Zhu’s research team has made significant contributions to the emerging field of federated analytics. They have developed several federated analytics algorithms and systems, addressing data analysis scenarios with privacy requirements, including data heterogeneity measurement, frequent pattern mining, and graph embedding. These research outcomes have been published in top-tier journals and conferences, such as the IEEE Journal on Selected Areas in Communications, IEEE Transactions on Mobile Computing, IEEE INFOCOM, ACM KDD, and IEEE ICDE.
