“Euclidean distance” is often used to measure those distances. Normalization gives equal weights/importance to each variable so that no single variable steers model performance in one direction just because they are bigger numbers.Īs an example, clustering algorithms use distance measures to determine if an observation should belong to a certain cluster. The short answer is - it dramatically improves model accuracy.

Now the question is how (on earth) exactly does this transformation help? Normalization is an essential step in data pre-processing in any machine learning application and model fitting. This process of normalization is known by other names such as standardization, feature scaling etc. The purpose of normalization is to transform data in a way that they are either dimensionless and/or have similar distributions. The other problem - which is not obvious but if you take a closer look you'll find it - is the distribution of data, which is quite different in these two variables (both within and between variables). One obvious problem of course is that these two variables are measured in two different units - one in hours and the other in miles. 5, 10, 25 hours ) and distance in miles (e.g. Let’s say we have a dataset containing two variables: time traveled and distance covered. Let’s first define what exactly is normalization. But of course, that wasn’t the focus of that article, so today I want to pick up on that. I touched upon how data normalization impacts clustering, and unsupervised algorithms generally.īut I felt that I missed the opportunity to go into more details. I wrote about cluster analysis in the previous article ( Clustering: concepts, tools and algorithms), where I had a short discussion on data normalization.

Sound normalizer software training