Interdisciplinary Training in
Complex Networks and Systems

Abstract

This talk reviews the work of our group on the problem of inverse classification, i.e., the perturbation of a test case to minimize its posterior probability of an undesirable class label, such as the predicted onset of a disease. Starting from exhaustive search on a k-nearest neighbor classifier, we develop mathematical optimization models to handle both smooth classifiers (e.g., SVMs) and general non-smooth classifiers (e.g., random forests). The ideas are further extended to causal modeling framework with a deep learning ANN. Results are applied to a recommendation system for patient risk minimization, incorporating a realistic and customizable cost model.

Biography

Nick Street is the Henry B. Tippie Research Professor and Departmental Executive Officer in the Management Sciences Department at the University of Iowa, with joint appointments in the Computer Science Department, the College of Nursing, and the Interdisciplinary Graduate Program in Informatics. He is also the director of the interdisciplinary graduate program in Health Informatics. His research interests are in algorithmic approaches to machine learning and data mining, particularly the use of mathematical optimization in inductive learning techniques. His recent work has focused on ensemble construction methods, knowledge transfer, correlation analysis, personalized health care decision making, and social network analysis. He has published over 120 journal, conference and workshop papers, and is the prior receipient of an NSF CAREER award.

CNS-NRT Student Discussion Seminar

• Date/Time: February 14th @ 1pm
• Location: Informatics East, room 226B

Papers to be discussed

• Lash, M.T., Lin, Q., Street, N. & Robinson, J.G. (2017) A budget-constrained inverse classification framework for smooth classifiers. In IEEE International Conference on Data Mining Workshops.