Skip to main content

in this section

ECE Ph.D. candidate wins best student paper award

Thursday, July 7, 2016

Igor Labutov, an ECE candidate advised by Assistant Professor Christoph Studer, has won the best paper award at the Ninth International Conference on Educational Data Mining (EDM) for his paper, Calibrated Self-Assessment.

In recent years, Massive Open Online Courses (MOOCs) have shown that education can be scaled far beyond the size of physical classrooms. However, these "first-generation" MOOCs have highlighted a major challenge—efficiently grading and providing feedback in classes with thousands of students.

The paper proposes a novel method for highly scalable assessment (i.e. grading) based on the idea of students grading themselves. They use a principled statistical method that models the students' behavior in order to compensate for individual biases, yielding calibrated estimates of the students' true grades that an instructor can rely on.

Igor’s research focuses on developing scalable machine learning models for interactive applications, primarily aimed at problems in education. Before Cornell, he received his B.S. in Electrical and Computer Engineering from The City College of New York. In the fall, he will begin a post-doctoral position at the Machine Learning department at Carnegie Mellon. His long-term goal is to develop the next generation of interactive learning tools that leverage the entire web.

 The EDM conference is a leading international forum for high-quality research that mines large data sets in order to answer educational research questions that shed light on the learning processes. These data sets may come from the traces that students leave when they interact with learning management systems, interactive learning environments, intelligent tutoring systems, educational games or when they participate in a data-rich learning context. The types of data therefore range from raw log files to eye-tracking devices and other sensor data.

 Find more information:

back to listing