Repository for the course "Mathematics in Machine Learning" at Politecnico di Torino.
In this project I had to analize the Dry Beans dataset with classic machine-learning methods. After a basic exploratory data analysis, I performed dimensionality reduction and applied resampling methods to balance the class distribution. Finally, I tried several classification algorithms with cross-validation combined with different data preprocessing and compared the results.
In the Jupiter Notebook there is a theoretical explanation next to every used method.