This is one the most interesting project I had to do in my Master degree.
This is a datamining project.
A data scientist from google came to the university to and introduce and teach us the basis of data science. The purpose was for us to choose a subject. I chose to do a soccer games predictor. I was a little bit naive at the beginning and though I could have good results with a few data.
On the beggining, I was kind of lost. This was all new to me. I had to struggle, talk with classmates, have big discussions. I also read books of data science. Thanks to this, I started to understand better how to conduce a data science project.
I first dealt with the data. This was the most difficult part, and I could not find enough data to get good results (I would have to get all the data from all the players for all the games since 10 years ago). I tried to scrap the website of french football league but I did not had much time left.
This is a classification problem, either you have the victory of the home team, either the victory of the away team, either draw.
I chose to use Random Forrest because I thought it would be efficient and prevent from overfitting.
Finally, I would say that this made me want to learn more and more about data science, and I would like to pursue this way.
The most interesting in this folder is the report. It is in french though.