NBA Hall of Fame Predictor




Data Science Project



About This Project

The primary purpose of our project was to determine whether statistics from a current NBA player’s early career has an indication on hall of fame induction. Using various machine-learning classifiers such as nearest neighbor, support vector machines, and decision trees, we analyzed twenty-seven relevant statistics from a large list of NBA all stars’ first four seasons and predicted whether or not a given player will be inducted into the hall of fame on this basis.

Project Data

Rahul Matta, Sachin Lal
Programmer
March-June 2015
Project Website




THE GOAL

On the whole, teams that win the NBA championship have at least one future hall of fame inductee.Given this fact along with the one that it is the primary goal of teams to win the NBA championship, it is absolutely in the best interest of teams to know an accurate prediction of a player's future.


THE RESULT

We were successful in training a model that can predict with 81% accuracy if a player will be in the Hall of Fame after their first four years in the National Basketball Association.


THE IDEA

We decided to train several classifiers; we trained Zero-R, Decision Trees, K Nearest Neighbor, and Support Vector Machines. We found that K-Nearest Neighbor worked best when k=10.

Takeaway

In this project I learned to work with big data as we trained a model to predict if a player would be inducted into the Basketball Hall of Fame based on their statistic in their first four years. We learned about the process to clean data so that we could take advantage of different models.

Send Me A Message:

sachinlal2016@u.northwestern.edu


Call Me:

781-864-3248


I'm Social


Created by Sachin Lal