Date Approved

9-19-2023

Embargo Period

9-25-2023

Document Type

Thesis

Degree Name

Master of Science in Data Science

Department

Computer Science, Mathematics

College

College of Science & Mathematics

Advisor

Umashanger Thayasivam, Ph.D.

Committee Member 1

Gregory Ditzler, Ph.D.

Committee Member 2

Shen Shyang Ho, Ph.D.

Keywords

decision science, automated decision-making, interpretable decision-making

Subject(s)

Machine learning

Disciplines

Computer Sciences | Data Science | Mathematics

Abstract

This abstract explores two key areas in decision science: automated and interpretable decision making. In the first part, we address challenges related to sparse user interaction data and high item turnover rates in recommender systems. We introduce a novel algorithm called Multi-View Interactive Collaborative Filtering (MV-ICTR) that integrates user-item ratings and contextual information, improving performance, particularly for cold-start scenarios. In the second part, we focus on Student Prescription Trees (SPTs), which are interpretable decision trees. These trees use a black box "teacher" model to predict counterfactuals based on observed covariates. We experiment with a Bayesian hierarchical binomial regression model as the teacher and employ statistical significance testing to control tree growth, ensuring interpretable decision trees. Overall, our research advances the field of decision science by addressing challenges in automated and interpretable decision making, offering solutions for improved performance and interpretability.

Share

COinS