The project is centered around discovering models for Massive Online Open Courses (MOOCS) — like MITx — to help assess and optimize course quality. It involves drawing data harvested from the MITx courses, formally structuring the data such that it highlights features of interest, generating models that provide us with useful insights, and applying these models to better understand and improve course quality. I will focus on scalability. Volunteer Compute offers a means of harnessing free, idle computing power on devices around the world and distributing computational load over the devices, so I will develop a framework on Volunteer Compute for running the learning algorithms to generate the models we need

I’ve been interested in data sciences for some time, and the introductory Machine Learning course I took last year got me hooked on the field. Since then I’ve taken on some projects that have further exposed me, such as my summer internship at Google, during which I’ve worked closely with the team that implements and supports an internal machine learning library, and a personal project dealing with Natural Language Processing.