Faculty Advisor

Beck, Joseph E.

Abstract

This project explores an approach for analyzing problem level data received from an intelligent tutoring system, ASSISTments. Through data processing techniques, a dataset representative of student answering patterns is constructed. This data is fed into various machine learning algorithms to model student competency. The output from one such algorithm, an LSTM neural network, is extracted to generalize across success metrics, which the original model was not built to predict. Such a model could be used to determine a threshold for student competency and detect when students need help early. Instructors can then act on this information and follow through with prevention techniques before the student fails.

Publisher

Worcester Polytechnic Institute

Date Accepted

August 2018

Major

Computer Science

Project Type

Major Qualifying Project

Accessibility

Unrestricted

Advisor Department

Computer Science

Share

COinS