Software Requirement Specification.md

Software Requirement Specification

NBA Match Predictor (version 3)

Implemented using: Python Eve. , MongoDB, Beautiful Soup, ...[insert more as project progresses]

Specifications:

1. Set up backend server using MongoDB to store team statistics

   • Deadline: 11/25/2019
   • Set up endpoints
   • Set up endpoints' schemas

2. Implement scraping function using Beautiful Soup:

   • Deadline: 11/28/2019
   • Ability to scrape team statistics on NBA Reference, Team Ranking
   • Ability to scrape game schedule on a specified date
   • Ability to scrape individual statistics in a specified season
   • Ability to scrape matchup's rosters for a specific game in the past

3. Have a professional U.I/ Interactive Front End

   • Deadline: 12/12/2019
   • Implement own frontend using HTML/CSS/Javascript
   • Host live website on an actual online server. Look into Docker, Heroku, Digital Ocean.

4. Implement a Neural Network to predict matchup winner, by training from individual players ' statistics:

   • Deadline: 12/20/2019
   • Scrape games from the past 5 seasons. Store in some endpoints ?
   • Train using matchup's rosters and individual players ' stats ?
   • Use own Neural Network implementation or consider TensorFlow, Scikit-Learn

5. Set up accounts/ email-list (Optional):

   • Deadline: 12/27/2019
   • Send registered users email of predictions.
   • Store user logins, authentication,.. somehow
     • Look into hashing function for password storage
     • Set up auto-mailing server (?)