34th Annual Student Research Conference
Using Deep Q-Learning to Predict Student Success
Samuel J. Myers
Dr. Ruthie Halma, Faculty Mentor
The ability to predict a student’s academic success can have major ramifications throughout their career. However, with increasing class sizes and online instruction, the ability for teachers or other school personnel to do so accurately is limited. Thus, it is worthwhile to investigate whether an artificial intelligence (AI) system can accurately predict student scores utilizing data on the student’s living situation, as well as quantify how the amount of data used impacts prediction accuracy. This project implements a supervised-learning AI system using Deep Q-Learning to analyze three data sets.
Keywords: Artificial Intelligence, Machine Learning, Neural Networks, Deep Q-Learning
Topic(s):Computer Science
Presentation Type: Face-to-Face Oral Presentation
Session: 102-4
Location: SUB Activities Room
Time: 9:15