I took part in Kaggle’s NLP competition called Disaster Tweets. The goal was to predict if given tweet is about real catastrophe or not.

Kaggle is online community of data scientists and machine learning engineers. There are many competitions and challenges you can take, and I decided to go with the one connected with NLP (natual language processing).

The best result was obtained using BERT, state-of-the-art language model developed in Google. At the moment of writing it allowed me to take 335th place out of 2698 participants.

I also tested multiple popular algorithms and ways of pre-processing data. The results, with actual code and descriptions, are available in my Github repo.

I encourage you to explore both competition and my solutions - it was a fun ride and hopefully you can find it interesting too.