This project aims to dive deeper into extracting the data in the GDELT Project which is a 'global database of society'. More specifically, I aim to investigate nationalistic movements around the world through NLP practices as well as investigating trends over time using time series analysis.
What will be challenging is determining what constitutes as nationalistic media since, most nationalists don't tend to use the word 'nationalism' when referring to themselves or their movement. So I will need to investigate ways to categorise articles as nationalistic. To do this, I might first start with training models on articles that are known to be nationalistic to determine patterns. The hard part will be extracting these pieces from the database.