Statistics, Machine Learning and Biology
Professor Shen works on general problems in statistics and machine learning, including, Bayesian models, nonparametric methods and biostatistics. “I focus on applications in biology, neuroscience, clinical trials, environmental and social studies,” he says. For example, he has collaborated with biologists and physicians to help develop new therapeutic options to treat heart diseases and to better predict the risk of developing liver cancer using information from biomarkers and other patient information.
Professor Shen is also exploring applications in sports analytics. “Recently, I have been working on developing new learning methods for analyzing data from sports games,” he says. In one project, he and his collaborators have developed a new Bayesian nonparametric clustering approach for analyzing the shot selection pattern data of NBA basketball players.
Bridging the Gap
Overall, Professor Shen’s goal is to promote the appropriate use of meaningful statistical methods in the world of science, thereby bridging the gap between theory and practice of data science. By promoting collaborations between mathematicians and statisticians and other domain experts, he works to develop flexible, efficient and robust statistical methods for modeling complex data. “A good model can be generalized to address other problems — that’s the beauty of statistics,” he says. “It lets you play in everyone’s backyard.”