Getting to Know U
The first goal of the project is to curate the large volumes of data collected by higher education institutions. The team will then develop dashboards and other tools that enable users and decision makers to analyze, understand, and act on that data.
Note: An identical VIP is being created at Georgia Tech, and our plans are to collaborate with that group (see https://www.vip.gatech.edu/teams/VXG).
Issues Involved or Addressed
Universities already collect large amounts of data from students (e.g. application data, background, GPA), faculty and staff (e.g. demographics, employment, awards), infrastructure (e.g. square footage, physical plant, power usage), finance entities (e.g. budgets, pay), research operations (e.g. awards, expenditures, patents), and the registrar (e.g. courses, programs, degrees). Such data is, however, siloed and not available as needed in a usable way for envisioning the status of the university. The issues involved in attempting to extract information from the data range from the regulatory and compliance (e.g. FERPA) to accuracy and consistency. The project will focus on building tools that progress from descriptive analytics (dashboard of historical data), to predictive analytics (predict future trends), and ultimately to prescriptive analytics (perform what-if analyses).
Methods and Tech
- Machine learning
- Data analytics and visualization
- Programming languages, e.g., Julia, Python, R and D3
Academic Majors of Interest
Open to all majors, but a particular focus on STEM majors that provide any emphasis on analysis.
Preferred Interests and Preparation
Skills: Students with undergraduate- or graduate-level standing and a willingness to work with large data sets and analysis tools
Attributes: Inquisitiveness, willingness to learn new tools and skills