This video explains how to properly evaluate the performance of unsupervised clustering techniques, such as the K-means clustering algorithm. We set up a Python example using the iris data set from scikit-learn to demonstrate the difference between classification and clustering problems, understanding why the permutation of cluster labels is an essential task. We also discuss two important measures, i.e., homogeneity and completeness. This video also provides an overview of performance evaluation methods for clustering algorithms, namely external and internal measures.
Review of three important clustering algorithms by Dr. Data Scienece: • Three Clustering Algorithms You Shoul...
#Clustering #ModelEvaluation #ClusterLabels