How does feature scaling affect distance-based machine learning algorithms, and how would you implement it using Scikit-learn?
