Predictive Analysis for High-Dimensional Data Sets

COMP 790-201

January 12 – February 11, T/TH 9:30-10:45 AM

 We will cover multiple techniques for each of four problems: sparse regression, classification, clustering, and dimensionality reduction. The only graded assignments in the class will be four short projects that apply the techniques presented in class to real data. Our focus in the course will be on understanding the basic idea behind the methods and using existing implementations, so it should be a gentle introduction even for those with no background in statistics or machine learning. The projects themselves are drawn from a wide range of data types and biomedical applications, including audio data from patients with Parkinson’s disease, image data from tumor biopsies, and gene expression data from cancer cells. By Joshua Welch and Robert Corty (both funded BD2K trainees), Jans Prins (Computer Science), and Dirk Dittmer (Microbiology and Immunology).

Leave a Reply

Your email address will not be published. Required fields are marked *