Predicting Student Retention using Machine Learning

Winner of the Dr. Chaz Schlindwein Research Award

Scott Gangloff and Joshua John

Our project examines the possible application of machine learning technology to aid university administrators in identifying at-risk students. Students will be classified into two categories: good standing (graduated or returned), or bad standing (i.e., transferred, dropped, or suspended). We have trained multiple different models to find the best performing one. Along with the classification task, we also identify the key attributes from our dataset that are the most significant factors in determining at-risk students. In conclusion, we will show the accuracy of our model at determining whether a given student will be at risk of dropping out or being suspended. Also, we aim to create a GUI (Graphical User Interface) that will allow university administrators to upload a new student data set and predict the at-risk students using our training model.

Scott Gangloff graduated from Sweet Home High School in Buffalo, New York. He is currently a senior in Computer Information Systems with an emphasis on Software Development. His experiences include working as a student research intern for Matt Braaten at Lander University and presenting at the CCSC : SE student research competition.

Joshua John is a senior Computer Information Systems major at Lander with an emphasis in Networking and Software Development. He is minoring in Cybersecurity and is currently the co-captain of the Cyber Defense Team and has competed in multiple collegiate level cybersecurity competitions. After graduation he plans to attend grad school in cyber security and machine learning. He is currently working as a student research intern for Matt Braaten at Lander University.