What is K-means Clustering?
K-means clustering uses characteristics rather than pre-defined categories to categorize unlabeled data in an unsupervised manner. The number of groups or categories produced is indicated by the variable K. The objective is to divide the data into K distinct clusters and give the location of each cluster's center of mass. A cluster (class) can then be given to a fresh data point based on the closed center of mass. This method has the major benefit of eliminating human bias from the analysis. The machine constructs its own clusters rather than having a researcher do it using empirical evidence rather than conjecture.