This repository is dedicated to performing a variety of NLP experiments on the COVID-19 Kaggle dataset.
The Geographic NER iPython notebook is focused around extracting geographic locations (cities and countries) mentioned throughout the papers.
The notebook is split into the following sections:
- Initial GeoTopicParser experiments.
- GeoText extraction (the actual location extraction).
- Weather extraction (from the extracted locations).
- Text processing to determine the severity of incidents in the papers per mentioned locations (currently in progress).
All the resulting data is saved in the geodata folder.