CS174 Project Proposal
Smart NBA Outcome Prediction System
Fanghao Chen
Introduction:
It is widely known that NBA (National Basketball Association) is the most popular basketball
league on the world. There are millions of fans of each team and they all wish their supported
team will be the season's champion. Here I propose a smart system to predict the outcome of a game
and finally a champion.
Dataset:
All the NBA data could be found in www.databasebasketball.com.
The database contains the year-long NBA data of players, teams and coaches for both regular seasons and playoffs.
Approach:
Given two teams, predict the outcome of this game by selecting several features from the data set.
The three candidate algorithms for classification are Neural Networks, Logistic Regression and KNN. Finally conduct an accuracy
comparison among those algorithms.
Timeline:
- 4/16: Research the literature and learn the selected algorithms.
- 4/22: Check and clean dataset through cross-validation.
- 4/26: Implement Neural Networks in Matlab.
- 5/8: Implement Logistic Regression in Matlab.
- 5/15: Implement KNN in Matlab.
- 5/22: Conduct the comparison across the algorithms.
- 5/25: Analyze the results.
- 5/30: Finish the write-up.
Reference:
[1] Colet, E. and Parker, J. Advanced Scout: Data mining and knowledge discovery in NBA data. Data Mining and Knowledge Discovery, Vol. 1, Num. 1, 1997, pp 121-125.
[2] McMurray, S. (1995). Basketball's new high-tech guru. U.S. News and World Report, December 11, 1995, pp 79-80.
[3] Balla, Radha-Krishna. Soccer Match Result Prediction using Neural Networks. Project report for CS534.
[4] Alan McCabe, An Artificially Intelligent Sports Tipper, in Proceedings : 15th Australian Joint Conference on Artificial Intelligence, 2002.
[5] Michael Papamichael, Matthew Beckler and Hongfei Wang, NBA Oracle, 2009.