This repository was made as an assignment for Fontys, where we had to create any type of 'media'. I chose the tough road and decided to create an LSTM neural network that generates tweets. For this project I used a dataset of Donald Trump's tweets, since there was already a present scrape, which drastically decreases the data gathering and cleaning process.
This network is able to generate Donald Trump tweets (poorly), based on a seed, which is a random word chosen from all of his tweets. Based on this seed the network will look for the most fitting words and try to replicate something that Donald Trump could have said.