2. 7. Data transformation and feature scaling

In data science and machine learning, data transformation is the process of modifying data into a more useful or suitable format. This could involve converting data types (e.g., strings to dates), combining columns, or applying mathematical functions to make data cleaner and more relevant. One common form of transformation is feature scaling, which adjusts numeric features so they are on a similar range or scale. Feature scaling is essential when working with algorithms that calculate distances or rely on gradient descent (like KNN, SVM, or logistic regression), because having one feature range from 0–1000 while another ranges from 0–1 could mislead the model’s priorities.