Speaker
Xin Zan, Assistant Professor, Department of Industrial and Systems Engineering at the University of Iowa
Abstract
In modern industry, a profusion of structured, semi-structured, and unstructured data has been emerging from diverse complex systems across many application domains, which provides substantial opportunities for harnessing data analytics to enhance intelligent decision-making. However, the opportunities coexist with new and significant data challenges for efficient analysis that make it challenging to extract useful patterns and meaningful insights, including: (i) data heterogeneity and structural complexity, arising from the multitude of data sources and formats; and (ii) limited availability of high-quality data, which is often subject to weak supervision during data collection and labeling processes in the context of a range of practical resource constraints in data acquisition, communication, storage and analysis, and human labor.
This talk concentrates on structured data analysis with weak supervision to develop systematic data-driven analytics methodologies for complex system modeling, monitoring, and diagnosis with applications in healthcare. The talk mainly covers two works and the extensions applied in public health surveillance and disease diagnosis: (i) data-driven resources allocation strategies to optimize the allocation of limited diagnostic tests for quickly detecting outbreaks of infectious diseases, despite insufficient or incomplete testing data among communities at the early stages; (ii) novel weakly supervised learning techniques for diagnosing sleep apnea by using massive signal data only with limited annotations, which significantly reduces manual efforts for data labeling in clinical practice.
Bio
Xin Zan is currently an assistant professor in the Department of Industrial and Systems Engineering at the University of Iowa. Xin received her B.S. degree in industrial engineering from Shanghai Jiao Tong University, China, in 2019, and her master’s degree in statistics and Ph.D. in industrial and systems engineering from University of Florida in 2022 and 2024.
Xin’s research focuses on advanced data analytics and system informatics to develop novel data-driven methodologies for complex system modeling, monitoring, and diagnosis, with a particular interest in resource constrained systems with weak supervision. Xin’s interdisciplinary research leads to immediate applications in healthcare, manufacturing, and service systems, etc. Her works have been recognized by multiple awards, including the Third Place winner of the CIC student paper challenge, Honorable Mention of INFORMS QSR Best Student Poster, Finalist of IISE QCRE Best Track Paper, and feature article in AIE.