This is a project to personalize academic advising with the goal of better identification of student risk and success in retention and graduation in higher education. The personalization is accomplished through the development of an interactive system using a novel approach to data modeling. The advisor is given the sequence of courses, labs, and sections that the student has taken, along with grades and any other assessment data that is available. This information is then fed to a data mining application that identifies the degree of success that the student has had and a prediction of any additional assistance the student may require in order to achieve academic success. The data mining application uses machine learning technology and interactive input from faculty, advisors, and academic leadership to accurately model student achievements.
More precisely, the interactive framework enables the discovery of actionable knowledge to improve student success by including the domain experts in data-driven discovery and decision-making from heterogeneous and longitudinal student data. The approach is to integrate and iterate the feature extraction, analytics, and interpretation processes within a single interactive user experience. Through the use of explorative interactive visualization of data and data patterns, the target user communities, including academic leadership, faculty, and advisors will be empowered to explore a broader range of meaningful hypotheses and derive specific actionable insights given the large and complex data that is being collected about students' performance and campus life. This will transform the ability to create policy, curriculum changes, and interventions that can address specific critical issues in universities more proactively than traditional analyses can provide for affecting retention, time to graduation, and student success.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
