In my understanding of feature scaling/normalization you need to do this, because of different measurement units inside the features and bring them to a uniform scale. And in my opinion this only belongs to data of different measurement units (ex. ft, quantity, kg, ...).
But I am unsecure if I can use data of the same unit of measurement (centimeters) without scaling, especially if I want to use distance measures.
The datasat which I use is the iris dataset (https://www.kaggle.com/datasets/uciml/iris) which consists of 4 features, but all of them are measured in centimeters. So do I need to scale or not?
The algorithm I want to use is the k nearest neighbor, a distance based algorithm.
