MM_2017

My implementation of various machine learning methods to predict the outcomes of the 2017 March Madness games

Data Sources

All data comes from Kaggle's yearly March Madness Competition

The data for 2017 can be found here: https://www.kaggle.com/c/march-machine-learning-mania-2017/data

Dependencies

• Python 2.7
• numpy
• sklearn

Features used

I didn't get a chance to really create many features for this data set due to a time crunch to finish before the tournament started. The features I used were:

• Offensive Quotient
• Defensive Quotient
• ELO Rating

Information about offensive and defensive quotient

http://a.espncdn.com/ncb/s/katzqa/010831.html

Information about ELO Rating:

https://en.wikipedia.org/wiki/Elo_rating_system

https://fivethirtyeight.com/datalab/introducing-nfl-elo-ratings/

Future Work

Put working in progress into usable methods

Create a tournament visualizer

Open to collaboration to improve this project