Effective Integration of Clustering and Classification or Regression Machine Learning Algorithms

This chapter includes a good exposition of centroid-based, density-based, distribution-based, and hierarchy-based clustering machine learning algorithms and various supervised machine learning models. K-means remains a popular unsupervised machine learning algorithm, we will discuss how the K-means can build hierarchical clustering. There are several classifiers and regressors in the literature and since the focus of the chapter is on unclassified, unlabeled, or unstructured datasets, the goal will be to start with clustering algorithm to determine the different classes or labels in the data. The resulting labeled data are used to build a classifier or a regressor as appropriate. This approach is important as significant datasets are unclassified or unlabeled. This approach has several applications, especially in natural language processing.