27-764 SPECIAL TOPICS: Data Analytics for Materials Science

6 units, Carnegie Mellon University

Materials Science and Engineering has traditionally been taught by emphasizing the development and application of technology. This course will present an alternative approach that combines data mining, data analytics, and material fundamentals i.e. materials informatics. Students will be introduced to informatics techniques related to data analysis. The mathematical basis of the algorithms will be explained. Examples in the course will be focused on materials problems. The main educational objective will be to enable students to use data analytics in their own research. The topics will include Examples of distributions found in materials science, e.g., lognormal, similarity of distributions, and p values Multiple Linear Regression with feature selection Principal Component Analysis PCA Canonical Correlation Analysis CCA Neural Networks and Computer Vision. There will be a short project in which students will apply appropriate techniques to a data set of their choosing. Students should be comfortable writing, compiling, and running simple computer programs in Python, R, or comparable environment. If not already familiar with python, students are strongly encouraged to engage in selfstudy before starting the course.

Prerequisites: none listed

Courseweave needs JavaScript to show sections, workloads and planning tools.