Statistical Learning and Data Mining in Engineering
ENGR 8130
3 hours
Statistical Learning and Data Mining in Engineering
Course Description
An exploration of engineering applications of statistical learning methods and techniques. Methods introduced include ridge, lasso, elastic net regression, logistic regression, principal components analysis, clustering methods, discriminant analysis, k-nearest neighbors, support vector machines, and tree-based methods. Artificial Neural Networks (ANN) and the recently emerged Deep Learning models are also introduced.
Athena Title
Statistical Learning
Equivalent Courses
Not open to students with credit in ENGR 8130E
Prerequisite
Permission of department
Semester Course Offered
Offered every year.
Grading System
A - F (Traditional)
Student learning Outcomes
Understand fundamental concepts and principles of different statistical methods and algorithms.
Choose proper methods for specific data structures or problems in engineering.
Be familiar with Python and R programming and computing environments (e.g., Anaconda and Jupyter notebook).
Conduct data analysis and modeling using Python scikit-learn package and R software.
Interpret the results in the specific contexts of engineering applications.
Topical Outline
Introduction and Statistics Review
Linear Regression (Ridge/Lasso/Elastic Net)
Convex Optimization
Logistic Regression
Principal Component Analysis
Clustering Analysis
Discriminant Analysis
K-Nearest Neighbors
Support Vector Machine
Statistical Learning Theory
Tree-Based Methods (Random Forest and Gradient Boosting)
Neural Networks
An individual term paper and an oral presentation is required. The term paper can be based on an ongoing research project or a new study initiated from the course. Students are expected to read and annotate the latest papers published in their respective fields of study. The annotated bibliography will serve as the basis for a literature review that supports the students’ primary research questions to be addressed by the term paper.
