Data Science in Architectural Engineering

Catalog description

This course aims to equip building science researchers and engineers with the tools for efficiently and effectively transforming data into knowledge, decision, and action. In particular, the course explores parametric and non-parametric regression methods, unsupervised clustering, classification, decision trees, random forests, and support vector machines. Additional advanced statistical learning methods are incorporated as they emerge and become relevant to the field. The course considers the multiple phases of statistical analysis, including data cleaning and pre-processing, exploratory analysis, model building and testing, data visualization, and reporting. Statistical learning topics are motivated through application and case studies involving thermal/building/renewable energy systems, energy efficiency, indoor air quality, and environmental engineering. This course provides a foundation for graduate researchers to engage in follow-on study of more advanced statistical learning methods, while also providing fundamental and beneficial data analysis skills to industry-bound students.