Master of Science (MS)
The Leadership and Career Accelerator (UNVS 101) is a course offered at San José State University (SJSU) designed to hone industry skills in and provide support to students of underserved backgrounds. The main goal of this study is to determine which features are most significant to identifying the students at risk of failing the course. This will allow faculty to better focus data collection efforts and facilitate an increase in classifier accuracy. The data came as three distinct sets (sources). One contained features describing student demographics and academic history, another described the students’ experience in the course, and a third showed the students’ readiness for the industry. The latter two were collected as survey responses from students. To score these features, three different machine learning (ML) techniques were used in two different schemes, although one of the schemes (principal component analysis a.k.a. PCA) was found to be ill-suited for this data set and for feature selection as a whole. The remaining scheme showed the survey features to be the most significant. ML models corroborated findings by testing specific feature subsets. Some overfitting was observed due to the data’s small size, but accuracies were consistently higher in models that were trained with the survey features. Additionally, a method employing clustering was developed that provided even more evidence of the survey features’ significance. This study ultimately proves that the UNVS 101 students’ survey responses are decidedly more important than their demographics and background when it comes to identifying their at-risk status. With continued diligence in data collection efforts, this study can be revisited using an expanded data set that will prove instrumental in more accurately identifying at-risk students.
Yesilyurt, Mustafa Emre, "Using Machine Learning to Maximize First-Generation Student Success A Contribution to the Mission of Aiding the Underserved" (2022). Master's Projects. 1198.
Available for download on Tuesday, December 19, 2023