Using Machine Learning to Detect Student Learning Levels along a Learning Progression

• In Event: Learning Progressions

Abstract

We introduced a novel method of using Natural Language Processing and Machine Learning to detect student learning levels using student short responses to a question measuring a learning progression of functions. Accuracy rate of our model was at 86%. Our precision, recall, and F1 scores ranged from .77 to .92.

Authors

• Duy N. Pham, UNIVERSITY OF MASSACHUSETTS, AMHERST

• Viet Lai, University of Oregon