Shluková analýza: Základní myšlenky a algoritmy

Abstract

The aim of this work is to introduce the reader to the issues of cluster analysis using practical examples and illustrations. The first and second chapters are focused on the description and analysis of the Iris dataset, which will be used during the work. The chapter devoted to cluster analysis begins with the formulation of the task and continues with the introduction of hierarchical clustering methods together with selected methods of non-hierarchical clustering (k-means, DBSCAN). The last part of the work is devoted to the clustering quality measures and application in finding the optimal number of clusters. For a better understanding, all methods are first described intuitively, then formulated with a mathematical apparatus, and then implemented in R.