NSF Funds Research for Improved Modeling

Project supports improvements for more robust computation and enhanced machine learning.

Dr. Xiaolei Huang, assistant professor in Computer Science, has received a CRII award from the National Science Foundation for his project "CRII: III: Metadata-guided Imbalance-Modeling for Robust Computational Healthcare." 

The project focuses on the idea that imbalance naturally exists in health data from text messages to electronic health records, which dampens the reliability, robustness, and trustworthiness of building computational healthcare models. This project proposes novel learning strategies that guide imbalance modeling by metadata and incorporate the varied imbalance patterns to promote model robustness. This project will examine and evaluate the proposed framework on a variety of health data by 1) new settings on different metadata factors and 2) effects and sensitivities of metadata factors for imbalance learning. Specific deliverables include developing a novel meta-learning toolkit with broad utility and educational activities to train the next-generation computational healthcare workforce.

This proposed research has significant transformative impacts threefold: interdisciplinary research, education, and outreach. As health data mining has many critical practical applications, improving robustness of health-related models will have broad impacts. Since little work has adequately addressed the imbalance-learning issue, there is a lot of room to make improvements that will benefit broad health fields that deploy machine learning models.

The educational impact is twofold. First, it will expose the graduate and undergraduate students (who themselves have diverse backgrounds in the PI’s lab) to the interdisciplinary field of computational healthcare. Second, the PI will integrate project outcomes into upper-division undergraduate and graduate courses, Introduction to Natural Language Processing and Advanced Machine Learning.

Finally, this project will utilize the unique opportunity to promote more reliable techniques for pediatric cancer research at the St Jude Children's Research Hospital. 

For more information on this project, contact Huang at xhuang7@memphis.edu.