The focus of this chapter is the discussion of basic clustering techniques.
- Basic concepts of clustering
- Similarity measures
- Hierarchical clustering methods
- Ward's hierarchical clustering method
- Nonhierarchical clustering methods
- K‐means method
- Density based clustering methods
- Model based clustering methods
After studying this chapter, the reader will be able to
- Discuss the various types of similarity measures and clustering techniques.
- Distinguish between hierarchical, non‐hierarchical and other types of clustering techniques.
- Perform cluster analysis for various types of data.
- Compare various clustering techniques.
- Summarize and interpret the cluster results.
- Use software packages MINITAB, R, and JMP to perform cluster analysis.
Grouping objects into one or more groups so that the objects within each assigned group are more homogeneous than otherwise is called clustering. Cluster analysis helps discover “natural” groupings of objects on the basis of similarities between the objects. Unlike classification methods, groups and the number of groups are unknown prior to clustering of data. Cluster analysis is an exploratory technique with no assumptions of group structure or number of groups and is often quite helpful to investigate the complex nature of data structures. Analysts could interpret and validate cluster analysis results based on their understanding of the data. Clustering can be achieved ...