INTERACTIVE CLUSTERING TOOLS IN MICROSOFT EXCEL PART 1. HIERARCHICAL CLUSTERING

Levon R. Hayrapetyan
Houston Baptist University

ABSTRACT

Hierarchical clustering is one of the most commonly used clustering methods in data
mining and involves creating clusters in predetermined, bottom-up order. The result is usually
presented in a dendrogram and most of software packages present the final dendrogram. In this
study, an interactive method is developed which visualizes each step of combining the two most
similar clusters into one and generates the corresponding subtree in the dendrogram. It also
allows “disassemble” the dendrogram, i.e., undo any (or all) clustering steps. The tool was tested
and have been successfully used in Data Analytics related course. It significantly improved
students’ understanding of the internal logic of hierarchical clustering.

Keywords: Hierarchical Clustering, Dendrogram, Data Mining, Interactive Tools