Machine Learning Course Paper: Predicting Coreference labels from Spoken Dialogue Tasks with Machine Learning: One-Hot Encoding and Feature Importances using Extremely Randomized Trees

Recently, I had to compete in a Kaggle competition for our Machine Learning for Data Science course at Columbia University. I used Sci-Kit Learn and some Python coding to execute my algorithm. I thought I’d share a diagram to illustrate the steps executed and the paper writeup.

The flow:

The full paper text: