CCEB Seminar-JASON H. MOORE, PhD, FACMI

"Accessible Artificial Intelligence for Epidemiologists" JASON H. MOORE, PhD, FACMI Professor Biostatistics, Epidemiology, and Informatics  Professor of Genetics  Director, Institute for Biomedical Informatics Jason H. Moore is a translational bioinformatics scientist, biomedical informatician, and human geneticist, the Edward Rose Professor of Informatics and Director of the Institute for Biomedical Informatics at the Perelman School of Medicine at the University of Pennsylvania, where he is also Senior Associate Dean for Informatics and Director of the Division of Informatics in the Department of Biostatistics, Epidemiology, and Informatics.  Moore's research focuses on the development and application of artificial intelligence and machine learning methods for modeling complex patterns in biomedical big data. A central focus is using informatics methods for identifying combinations of DNA sequence variations and environmental factors that are predictive of human health and complex disease. For example, he developed the multifactor dimensionality reduction (MDR)[1][2] machine learning method for detecting and characterizing combinations of attributes or independent variables that interact to influence a dependent or class variable. He then applied MDR for improved understanding of the interplay of multiple genetic polymorphisms of complex traits in genome-wide association studies. More recent work focuses on computational methods such as the tree-based pipeline optimization tool (TPOT)[3][4] for automated machine learning and data science. Current work also focuses on methods and software for accessible artificial intelligence.[5