AI
AI
Artificial Intelligence
Institutional Data & Analytics (IDA) has developed advanced machine learning models and generative AI agents to support data-informed decision-making across the university. Below is a portfolio of AI-powered solutions currently developed and maintained by IDA:
AI Agents
IPEDS Agent - Provides precise, institutionally validated information based on official data submitted to Integrated Postsecondary Education Data System (IPEDS). It serves as the authoritative source for officially reported figures across enrollment, admissions, retention, graduation, finance, financial aid, and human resources aligned with IPEDS definitions.
External Survey Variance Agent - The External Survey Variance Agent compares current-year external survey submissions with prior-year filings to identify notable year-over-year changes. By flagging unexpected shifts that may require review, the agent supports data accuracy and consistency.
Predictive AI/Machine Learning Models
FTU First Year to Second Year Retention Model - This predictive AI model estimates the likelihood that first-time, degree-seeking undergraduate students will return for their second year. The model enables early identification of at-risk students for targeted advising and intervention.
Rowan Online - Term to Term Retention Model - The Rowan Online Term-to-Term Retention Model predicts whether a Rowan Online student will persist from one term to the next. It identifies students at risk of stopping out early in the term to support timely advising and outreach.
First-Time UG Enrollment & Scholarship Model - This predictive and optimization model estimates enrollment probabilities for first-time, degree-seeking undergraduate students and evaluates the impact of scholarship offers on yield.
UG Students Forecast Model - This model forecasts Regular University undergraduate enrollment across academic periods. It estimates both retention and new student headcount and compares projections to actual outcomes. The framework helps evaluate forecasting accuracy and identify variance trends across student categories.