Turkish Soccer League Prediction
This project was attempted during an internship and turned out to be not so promising. As I was not able to collect enough data to attempt any form of deep learning nor did I have enough data for a classical machine learning algorithms. My hope is now to attempt the same procedures to another soccer league with more publicly available data. My results were better than random or better than just saying the home will win each game(this is the 45% of the cases) this indicates that even with small data sets machine learning can be useful.
This project is not extraordinary, but it can be a good starting point for a bigger project or for someone who is very new to the field of machine learning. I hope it finds the person who it will be most use to in the ocean that we call the internet...