The present dissertation investigates elements of domain formalization, resource allocation, and student success in higher education to conceptually design a university-wide system to assist in strategic planning efforts. The proposed system is a program-level tool with a modular design to allow scalability and generalizability across the entire university. Higher education strategic planning decisions are under investigation by stakeholders and transparency is needed. University resources allocation models are often outdated lack to adequately support program-level decisions. Further, with the dynamic nature of technology, domain knowledge components are evolving rapidly. This complicates the situation as updating curriculum takes additional time and resources. Using the University of Central Florida's (UCF) School of Modeling, Simulation, and Training (SMST) as a case study to build and validate the system, I investigate Modeling and Simulation (M&S) domain knowledge, skills, and abilities (KSAs) using a series of natural language, text mining, and machine learning techniques to model topics within domain-specific texts including publication abstracts, job postings, and graduate course descriptions. From there, I use this information to identify and enumerate terms used to develop M&S ontology and expert models for the future university-wide system. This investigation benefits both the M&S field of study, clarifying ill-defined domain components and it helps inform the design of university-wide strategic planning systems.
Doctor of Philosophy (Ph.D.)
College of Sciences
Modeling and Simulation
Length of Campus-only Access
Doctoral Dissertation (Open Access)
Leis, Rebecca, "Natural Language Processing for Modeling Domains in Higher Education" (2020). Electronic Theses and Dissertations, 2020-. 376.