My data science portfolio
This portfolio is an ongoing project featuring my latest work and selected projects.
*Project links are disabled pending some updates and other changes*
- Collected and Analyzed Political 400,000 Tweets from 2007--2022
- Created function to pull tweets in 100 tweet increments allowed by Twitter APIv2
- Parsed and cleaned JSON data produced by API
- Created a corpus and document feature matrix for processing prior to analysis.
- Performed bag-of-words text analysis of trends by date and user.
- Pulled data from NHTSA's Fatality Analysis Recording System
- Cleaned and parsed federal data labels and accounted for sparse entries
- Conducted Poisson and Zero-inflated model regression
- Parsed and cleaned 27,812 responses to professional/demographic survey
- Converted currencies to USD and adjusted for inflation
- Performed spot-checks on data and accounted for issues found
- Created extensive visualizations and analyses on resulting data