Data Science Projects
This is a snapshot of projects I have taken on. Please note that I have worked with a variety of data types (text, images, numbers, financials...) to solve multiple problems.
App Prototype
Extracting data from credit card transactions in 4 different languages and automating data mining, analysis and visualizations
Outcome: Accuracy of solution exceeded client expectations by 36.4%
Tools: APIs, Streamlit, BeautifulSoup, Selenium
Named Entity Recognition
Identifying and extracting Surf Pros and Surf Locations mentioned in 39,552 YouTube video titles
Outcome: About 69.6% of the titles did not mention pro surfers or locations.
Tools: OpenAI API
Text Classification with LLMs
Tested and optimized various LLM models to classify >290,000 scientific publications using their abstracts
Outcome: Classification was 95.8% accurate.
Tools: NLP, SciBERT, BERT
Social Network Analysis
Assembling, analyzing and visualizing social networks among Fortune 500 corporations using their directors' board seats
Tools: networkX, iGraph
