Predicting Ballon d'Or winner with Machine Learning
Stack : R, Python (Scikit, Seaborn, Pandas)
- Scrapped 600 football players statistics (Goals, Assists, Red cards...) from Transfermarkt API
- Scrapped Ballon d'Or ranking from 1995 to 2019 from Wikipedia
- Merged and cleaned data to create the dataset
- EDA
- Tried predicting the 3 best scores (% of votes)
- Not a great success (overfitting with a MAE of 8%)
- Probably due to the lack of data and mistakes in data augmentation method