Is possibile to extract some relevant features from a dataset using K-Means? I know that K-Means is used to create some groups(clusters) of features in the dataset in an unsuper