PMO 1028 An Introduction to Machine Learning
Machine learning is a method of data analysis that designs models and algorithms that learn from data and make predictions based on data. Being a very active field, machine learning has been used in disease diagnosis, drug discovery, human genome research, and many other scientific areas. In this course, students will learn nine supervised classification methods: Bayes classifier, nearest neighbors classifier, classification through logistic regression, decision trees, bagging, random forests, support vector machines, LASSO, and neural networks. Students will learn two unsupervised learning approaches: K-means clustering and hierarchical clustering. Students will also learn deep learning. The statistical R language will be emphasized in the course. Helpful examples based on disease data will be presented. There are 10-11 three-hour long lessons in this course. For each lesson, the first two hours will be given to “theoretical” delivery of machine learning, while the last hour will be devoted to the “applied” lab with R. Upon completion of the course, students are expected to have a reasonable level in understanding the fundamentals of machine learning and mastering some of the most commonly used tools and techniques in machine learning.
Prerequisite
PMO 503 required,
PMO 504 preferred. In absence of
PMO 504, an approval from the course director is needed.