Data Science Projects
Predicting University Graduation Rates: Multiple Linear Regression Case Study
How can universities improve their graduation rates? In this project, I act as a data science consultant to identify the key predictors of student success using a Multiple Linear Regression (MLR) framework. To accomplish this, a predictive model of College Graduation Rate was developed using quantitative university data.
Climate Data Analysis in Python & Jupyter Notebook
In this project, climate data analysis was performed for analyzing sea-ice and ocean state estimates obtained from an MITgcm climate model of ocean circulation and sea-ice mechanics. The purpose was to uncover meaningful physical trends in the global ocean system in support of UT Austin's Oden Institute for Computational Engineering and Sciences ECCO research effort. Results were compared to those obtained from well-cited climate models in the literature and to theoretical behavior. This was accomplished using scientific computing tools in Python and Jupyter Notebook.
Case Study: Biases in Facial Recognition Technologies
The purpose of this project was to examine the causes and impacts of racial bias in facial recognition systems used by law enforcement, with a focus on how biased training data leads to higher misidentification rates for racial minorities. It discusses both technical and non-technical solutions, including improving dataset diversity, advancing bias-mitigation algorithms, implementing legal restrictions, and increasing diversity within engineering teams.
Capstone Project: Automatic Target Recognition for Search and Rescue Operations
Empower
Growth
The purpose of the project was to design and implement an Automatic Target Recognition and Map Generation system to support the Austin Fire Department in improving its search-and-rescue operations. The project focused on improving post-flood search and rescue by addressing the difficulty of locating distressed individuals on rooftops. A UAS-based computer vision system autonomously detected targets, determined their GPS positions, and generated maps for first responders.
Statistical Exploration in R: Performance, Risk, and Inequality
Using statistical analysis and data visualization techniques in R, these projects analyze Olympic athlete characteristics and performance patterns, investigate seasonal and demographic risk factors in Mount Everest expeditions, and assess long-term trends and disparities in U.S. median household income across racial groups. Collectively, the projects demonstrate the ability to clean and structure large datasets, apply exploratory and multivariate analysis methods, and extract meaningful insights from diverse domains.