Implementing clustering using Python

Now, as we understand the mathematics behind the k-means clustering better, let us implement it on a dataset and see how to glean insights from the performed clustering.

The dataset we will be using for this is about wine. Each observation represents a separate sample of wine and has information about the chemical composition of that wine. Some wine connoisseur painstakingly analyzed various samples of wine to create this dataset. Each column of the dataset has information about the composition of one chemical. There is one column called quality as well, which is based on the ratings given by the professional wine testers.

The prices of wines are generally decided by the ratings given by the professional testers. However, this can be very subjective and certainly there is a scope for a more logical process to wine prices. One approach is to cluster them based on their chemical compositions and quality and then price the similar clusters together based...