ABSTRACT
Grouping or clustering of objects basing on multidimensional data is a complex phenomenon but has potential applications in various areas like marketing research, psychology, facilities planning, drug discovery and gene expression analysis. Clustering of people based on socio economic features or food habits and education is a common issue in social science research and in community health studies. The tools used in cluster analysis are mostly computer intensive and a great deal depends on data visualization.
This chapter contains a discussion on methods of cluster analysis with focus on ability to handle a real problem. Two broad methods of clustering namely hierarchical clustering and k-means clustering are discussed in detail with the help of SPSS and R. The data setup and standardization are also focused. The dendrogram and its interpretation are explained in a lucid manner. Simple code in R to produce high quality output is also discussed. (148)
