Smart Data Analytics for Supervised Classification

MIT Chemical Engineering graduate student Fabian Mohr presented work on Smart Data Analytics, a research effort from the Braatz group in the Chemical Engineering department. The goal of the Smart Data Analytics project is to develop a systematic approach for determining the best predictive algorithm to apply to a given data set based on the properties of the data, such as nonlinearity, dynamics, and collinearity. Currently software has been developed for the regression problem, and Fabian presented two biopharmaceutical applications. Fabian also presented newer work developing the framework for classification prediction problems.