The app can be accessed at: https://ncaa-ml.herokuapp.com/

The purpose of this project is to:

1. Visualize NCAA Basketball and Football programs across the US
2. Analyze state budgets and how they may impact sports programs
3. Build Machine Learning models that predict:

• Which conference should you go to in order to be drafted by NBA?
• What is the probability of you getting drafted in the NFL?

4. Host a serverless app

• Extract and Transform
• Data Visualization
• Machine Learning
• Cloud

• Python
• Flask, Pandas, Jupyter
• PostgreSQL
• Google Colab
• Tableau
• Microsoft Azure ML
• HTML/CSS/Bootstrap
• Heroku
• AWS services: S3, RDS

• Gather data
• Read in data
• Data
  • Parse (multiple data files)
  • Map data files (implementing key relationships)
  • GeoCode Data (using Google Maps API)
• Load cleaned data into PostgreSQL connected through AWS RDS
• Create visuals using Tableau
• Construct a Machine Learning model using Microsoft Azure ML Studio
• Build a Flask app that renders our app with ML results on Heroku

• Initially when hosting with AWS, ran into difficulties connecting PostgreSQL to app.
• ML model request would time out on AWS when hosting.
• Building and optimizing ML models
• Web scrapping data

• Create a route on the app to allow user to access tables on PostgreSQL upon request.

• Sports-Reference
• NBA Draft Combine Data
• Athletics Budget Data

Team Leads: Salvador Olivas (https://github.com/solivas89) | Nabeel Sheikh (https://github.com/nsheikh23)