Projects
Independent and course projects with public code. You can find more in my portfolio.
Snopes Fact-Check Classifier
Scraped Snopes fact-check articles from January to July 2025, then built a TF-IDF and logistic regression classifier to predict whether a claim was rated true or false. The model reached 0.94 accuracy, with an F1 of 0.83 on the smaller “false” class. I also built a Tableau dashboard of article categories and ratings.
Tools: Python, Beautiful Soup, scikit-learn, Tableau
Code · Kaggle dataset · Dashboard
Political Contributions Database
Used the FEC’s API to scrape financial contributions to politicians from January 2022 to November 2024, and stored them in a queryable relational database.
Tools: Python, SQL, SQLite
NLP on Medical Company Earnings Calls
Group project testing whether sentiment and topics in earnings call transcripts could predict stock price changes. I cleaned the web-scraped transcripts and ran the sentiment analysis and topic extraction.
Tools: Python, NLTK
Topic Modeling Taylor Swift’s Albums
Used Latent Dirichlet Allocation on lyrics to test whether all of Taylor Swift’s songs really do sound the same. (They don’t.)
Tools: Python, gensim, spaCy, NLTK