Reference Work Entry

Encyclopedia of Machine Learning

pp 154-159

Categorical Data Clustering

  • Periklis Andritsos
  • , Panayiotis Tsaparas


Clustering of nonnumerical data; Grouping


Data clustering is informally defined as the problem of partitioning a set of objects into groups, such that the objects in the same group are similar, while the objects in different groups are dissimilar. Categorical data clustering refers to the case where the data objects are defined over categorical attributes. A categorical attribute is an attribute whose domain is a set of discrete values that are not inherently comparable. That is, there is no single ordering or inherent distance function for the categorical values, and there is no mapping from categorical to numerical values that is semantically meaningful.

Motivation and Background

Clustering is a problem of great practical importance that has been the focus of substantial research in several domains for decades. As storage capacities grow, we have at hand larger amounts of data available for analysis and mining. Clustering play ...

This is an excerpt from the content