This project contains code from the tutorial How to solve 90% of NLP problems: a step-by-step guide. The original data for the project can be downloaded here. This data is not formated correctly and has line folding. The formated data for the project with only relevant columns can be found on their repo and downloaded here. It contains 10,000 tweets.
The objective is try to predict tweets that are about actual disaster events based on the language in tweet. In particular, we want to distingush whether the tweet is talking about a joke or a movie review using language similar to disasters or refering to actual diasters.