As a Student Research Assistant at the Conrad Blucher Institute under Dr. Lucy Huang's mentorship at Texas A&M University-Corpus Christi, I focus on understanding COVID-19 vaccine hesitancy among diverse socio-demographic and racial groups in Nueces County. This research is crucial for identifying key factors influencing vaccine hesitancy and informing public health policies.My work involves in-depth data exploration and preprocessing. I use Principal Component Analysis (PCA) and Ordinary Least Squares (OLS) techniques to reduce dimensionality and uncover the data's underlying structure. These methods set the stage for accurate predictive modeling.I apply advanced machine learning and deep learning models to perform predictive analysis, including linear regression, logistic regression, kernel localized linear regression, XGBoost, and Graph Neural Networks. I rigorously evaluate and compare models based on precision values to ensure actionable insights.Data visualization is a key aspect of my role, where I use Matplotlib and Seaborn to create clear visualizations, such as histograms, bar graphs, and line graphs. These help illustrate data distributions, examine correlations, and present model outcomes effectively.To enhance model interpretability and precision, I employ SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-agnostic Explanations) techniques. These tools provide critical insights that can shape public health policies addressing vaccine hesitancy in Nueces County.

🚀 Exciting Project Update 🚀I recently had the opportunity to work on a project that combined computer vision and financial data analysis to enhance decision-making in stock trend analysis.Using a lightweight vision model, I developed an image processing system that extracts critical insights like Profit-Loss and divergence patterns directly from stock trend images. This approach not only streamlined the data extraction process but also increased the accuracy of financial data insights by 18%.To further optimize our workflow, I built a vector database using PostgreSQL (with psycopg2), specifically tailored to store and retrieve these stock trend images efficiently. By leveraging embeddings, we were able to reduce data retrieval time by 20%, making the system faster and more effective in analyzing market trends.This project was a rewarding experience in merging AI-driven vision models with data engineering to deliver real value in financial analytics. Looking forward to further exploring the potential of AI in transforming finance.#ArtificialIntelligence #ComputerVision #DataEngineering #FinancialAnalytics #MachineLearning #AI #PostgreSQL #StockAnalysis

Deploying custom software for a Proprietary Intellectual Property for research purposes - statistical analysis, literature and visualization.

Really! Grateful for this job, learned a new software technology for documentation- Laserfiche. Worked in the President's Office, it was a mix of fun and knowledgeable experience!