Yunlong (Mark) Wang
Machine learning on health data at Advanced Analytics, IQVIA in Wayne, Pennsylvania. I teach CSC 381/576 Data Science at West Chester University.
Research
A Patient World Model for Early Forecasting of Digital Health Campaign Outcomes: Capabilities and Limits
An intervention-conditioned world model that calls a digital health campaign's year-end prescription volume from its first four weeks — and, when the future media plan is unknown, imagines its own and forecasts just as accurately. With an account of where that capability stops.
Read the paper →Teaching · CSC 381/576 Data Science
Interactive modules for the course. Open one on any laptop or phone; nothing to install. Each runs on a dataset the class has already met.
Visualization
A dashboard on the 891 Titanic passengers. Filter by class, sex, port and age; every number and chart answers the same slice.
Inference
Sampling, confidence intervals and hypothesis tests you can re-run with a click.
Supervised learning
Regression and classification: change the model, watch the metrics move.
Unsupervised learning
Clustering and PCA on the same data, with the number of clusters in your hands.