During this 2-day long hackathon, we worked as a team to clean, visualise and analyse a large set of data regarding traffic accidents in the UK in 2015.
The aim of our project was to come up with some sort of model that would predict the class of severity of accident casualties. To do this, we thought of various models we could use, including using a Time Series Analysis using RNN, and using a Random Tree Classifier to make sense of the various features given to us in the data. We soon also realised that a major part of this task was cleaning up the data to remove noise and make it more analysable.
You can view our GitHub repository for the project here.