In this chapter, we learned the following:
Clustering is an unsupervised predictive algorithm to club similar data points together and segregate the dissimilar points from each other. This algorithm finds the usage in marketing, taxonomy, seismology, public policy, and data mining.
The distance between two observations is one of the criteria on which the observations can be clustered together.
The distance between all the points in a dataset is best represented by an nxn symmetric matrix called a distance matrix.
Hierarchical clustering is an agglomerative mode of clustering wherein we start with n clusters (equal to the number of points in the dataset) that are agglomerated into a lesser number of cluster based on the linkages developed over distance matrix.
K-means clustering algorithm is a widely used mode of clustering wherein the number of clusters need to be stated in advance before performing the clustering. K-means clustering method outputs a label for each row of data depicting...
