Meeting notes - Data from Kaggle - What data to use - Dynamic website vs different methods of machine learning
For this meeting I needed to make sure I could get the data from NBA games over the past seasons. I was successful in doing so by using Kaggle which contains NBA data dating back all the way to 1946. Ultimatly this is to much data so I need to go thorugh and decide how much of it I am going to use of it to make my prediction model. Also we talked about the additional steps to complete the project. There are two options that JJ and I talked about. One is making a dynamic website that will update daily with games and show predicted outcomes. Two is using multiple machine learning methods and fiding the best one.
Todo - Research machine learning models - Data clean up - Research dynamic websites
The first thing I need to do is sort through all the data and clean it up to only what I need. I also need to continue researching different machine learning models aside from sklearn. Finally I need to see how viable it is to make and use a dynamic website.