NBA_Result_Predictor

NBA score prediction system via neural network.

License

This project is licensed under the MIT License - see the LICENSE file for details

Develop environment

Python 3.5
Django 1.10 for data crawling server. We use csv data files for learning during preseason.
Tensorflow 1.2

Role

Data crawling : Using django and beautifulsoup, all game data are stored in csv files.
Learning : Using tensorflow, prediction will be executed.(Precisely we will use neural network) We won't use win-lose prediction, but score-to-score prediction.

Data source

All learning data files are crawled in 3 sources. [16-17 season only]

Standings : http://www.espn.com/nba/standings
Each team stats : http://www.espn.com/nba/team/stats/_/name/team_name/
Team by team comparison per game statistics : http://www.espn.com/nba/statistics/team/_/stat/team-comparison-per-game

Contributor

Data crawling
Park Seon Ha : Special Thanks
Yu Jeongmin
Jung Jo Hyung

Learning