Partition-based clustering with k-means

Author

Nielsen, Frank

Abstract

Nowadays, huge size data-sets are commonly publicly available, and it becomes increasingly important to eefficiently process them to discover worthwhile structures (or “patterns”) in those seas of data. Exploratory data analysis is concerned with this challenge of finding such structural information without any prior knowledge: in this case, those techniques that consist in learning from data without prior knowledge are called generically unsupervised machine learning.