Tweeted By @GaelVaroquaux
New paper: Similarity encoding for learning with dirty categorical variableshttps://t.co/j2JQxy5FcC
— Gael Varoquaux (@GaelVaroquaux) June 8, 2018
Code on https://t.co/9Saecz75MP
Useful for data scientists fighting with dirty data pic.twitter.com/0hmt0FXTfR