The public-facing version of a sentiment analysis tool built for my bachelor thesis, it uses the analyzers from TextBlob, VADER, and SiEBERT
Data: The good dataset from the Thesis was obtained at: https://www.kaggle.com/datasets/tumanovalexander/nyt-articles-data/code The bad dataset was obtained via API calls with the scraper provided in the code. A NYT API key will be required and can be generated here: https://developer.nytimes.com/apis
to install the environment run pip install -r requirements.txt on a Python install
If you should experience issues getting Cuda to work this site helps: https://pytorch.org/get-started/locally/