ML & survey science

Research-assistant work in the BYU Information Systems department: an end-to-end machine-learning pipeline and a validated survey study analyzed with structural equation modeling.

Two studies

A prediction pipeline, public on GitHub. An end-to-end machine-learning workflow in Python that compares six classifiers to predict Disney Lorcana game outcomes. An ensemble reached 0.73 AUC, evaluated with train/test splits, feature scaling, 10-fold cross-validation, probability calibration, and bootstrap confidence intervals.

A survey study on workplace technology health. I developed validated measurement scales across seven constructs, cleaned and prepared 350+ responses, and ran exploratory and confirmatory factor analysis and structural equation modeling to validate the model. Manuscript under review.