Research
Factors Influencing Trust in AI
As artificial intelligence becomes increasingly integrated into decision-making processes, trust continues to present a significant barrier to its adoption. This study examines trust across multiple domains including general use, employment, law enforcement, healthcare, and public policy. Using correlation analysis and Random Forest models, the study identifies the factors most strongly associated with trust in AI.
Predicting Flight Delay
Flight delays cost airlines billions of dollars annually and create significant inconvenience for travelers. This project applies machine learning models to predict flight delays using historical flight data, weather conditions, and aircraft characteristics. By identifying the key factors associated with delays, the analysis provides insights that can help airlines optimize scheduling and improve operational efficiency.
Universal Healthcare Analysis
Oregon healthcare officials conducted a policy experiment to assess the impact of expanded healthcare coverage on residents. As part of this study, my team and I analyzed patient health data to predict the likelihood of an emergency department visit. Using a logistic regression model, we estimated visit probabilities and identified the factors most strongly associated with emergency department utilization.
Seattle Collision Trends
The United States is experiencing a rapid increase in traffic volume. This study evaluates changes in vehicle collisions in Seattle, Washington, using time-series forecasting methods. By incorporating variables such as collision type, location, severity, weather conditions, and lighting conditions, the analysis identifies key factors associated with collision patterns.
Omaha Crime EDA
As the largest law enforcement agency in Nebraska, the Omaha Police Department operates with an annual budget exceeding $160 million and employs approximately 900 officers. Responding to more than 50,000 calls each year, this analysis examines criminal trends across the metro area and identifies regions with the highest frequency of reported incidents.