Giora Alexandron Gives Data Science Seminar Oct 26, Joint with Digital Learning and Hariri Institute for Computing

"Finding Instructional Resources that Assist Students: A Machine Learning Approach"

Since MOOCs contain both instructional resources and assessment items (questions), and also produce logs of learners’ interactions with these materials, they allow analyzing the relation between learners’ behavior in the MOOC, and their performance. This analysis can be used for making recommendations for future learners, and for giving feedback to instructors regarding the effectiveness of the course design.

In this talk, Giora Alexandron will concentrate on a model that relates students’ interaction with specific instructional resources (videos, e-texts, questions) to their success in solving specific questions. By running machine learning algorithms on this model, this work attempts to predict students’ success based on the resources that they have seen, and answer questions such as which resources are helpful for each question. He will describe the model and the methods, present some results, and discuss open questions and challenges.

About the Speaker: Giora Alexandron is a postdoctoral researcher at MIT. The focus of his research is applying data science tools to data from Massive Open Online Courses (MOOCs), with the purpose of developing engines for analytics and adaptive learning. He holds a PhD in computer science education from the Weizmann Institute of Science, and an M.Sc. in computer science from Tel-Aviv University. Between his M.Sc. and PhD he has spent a few years in the industry developing programming languages and tools.

This talk is co-sponsored by the Hariri Institute for Computing, Digital Learning Initiative, and Data Science Initiative.