Projects

Predicting Stroke with Synthetic Data

Machine Learning - Fall 2024 - Report and Presentation

Predicted stroke using a synthetic sample of electronic health records (EHR), meant to represent people ages 30-70 from Massachusetts using machine learning. As a data analyst working in education and health fields, I wanted to explore machine learning within the context of predicting outcomes in specific populations.

Reddit Vaccine Discourse: Cross-Vaccine (HPV, MMR, COVID), Cross-Time (2005-2023)

Independent Research - Argument Analysis - Fall 2024

Analyzed online vaccine hesitancy trends using natural language processing (NLP) to explore the “spillover effect” of COVID-19 vaccine attitudes on other vaccines, based on a Reddit corpus. This ongoing research is in collaboration with UMass researchers and is in the pre-publication stage.

Syringe Service Program (SSP) Evaluation Plan

Program Evaluation - Spring 2024 Program Evaluation

Developed a program evaluation for assessing the impact of Syringe Service Programs (SSPs) on reducing HIV rates and improving health outcomes for people who use drugs (PWUD), with a focus on data-driven decision-making for state-funded harm reduction programs.

Maine’s Unified Data Warehouse (MUD) is a fictional warehouse which I piloted in my time working at Maine’s Office of Behavioral Health. This evaluation was not commissioned by the State of Maine

Exploring Relationships and Centralities in Healthcare Networks of People Who Use Drugs (PWUD)

Social Networks - Spring 2024 - Poster

Explored healthcare journeys of people who use drugs (PWUD) using synthetic electronic health records (EHR), mapping their patient, provider relationships.The average recovery journey for a PWUD lasts over ten years and involves 6 treatment episodes. This project serves as an introduction to understanding patient-provider encounters of PWUD and as a template for future analysis.

A Sentiment-Based Exploration of Glassdoor Reviews in US Companies with Unlimited PTO Policies

Text as Data - Fall 2023 - Poster

Conducted sentiment-based text analysis to compare employees’ written reviews with their star ratings, evaluating alignment and satisfaction regarding paid time off.