Skip to main content

Daniel Marley

DHTS Developer
Duke University

Research Interests


I am currently interested in using graph neural networks (GNNs) to address problems in healthcare.  GNNs offer a flexible solution to many problems that require context and information from a variety of sources to reliably solve.

My research interests build on my recent work in machine learning
  • Stöckl, M., Seidl, T., Marley, D., & Power, P. (04 2021). Making Offensive Play Predictable -Using a Graph Convolutional Network to Understand Defensive Performance in Soccer.
  • Stats Perform, System and Method for Evaluating Defensive Performance using a Graph Convolutional Network, US Patent Application No. 17/649,970
  • Stats Perform, Action-Actor Detection with Graph Neural Networks from Spatiotemporal Tracking Data, US Patent Application No. 17/645,539
  • Stats Perform, Graph-based Method of Predicting the Next Pitch, US Patent Application No. 17/226,211