Categorization - Conceptual Clustering

Conceptual clustering is a modern variation of the classical approach, and derives from attempts to explain how knowledge is represented. In this approach, classes (clusters or entities) are generated by first formulating their conceptual descriptions and then classifying the entities according to the descriptions.

Conceptual clustering developed mainly during the 1980s, as a machine paradigm for unsupervised learning. It is distinguished from ordinary data clustering by generating a concept description for each generated category.

Categorization tasks in which category labels are provided to the learner for certain objects are referred to as supervised classification, supervised learning, or concept learning. Categorization tasks in which no labels are supplied are referred to as unsupervised classification, unsupervised learning, or data clustering. The task of supervised classification involves extracting information from the labeled examples that allows accurate prediction of class labels of future examples. This may involve the abstraction of a rule or concept relating observed object features to category labels, or it may not involve abstraction (e.g., exemplar models). The task of clustering involves recognizing inherent structure in a data set and grouping objects together by similarity into classes. It is thus a process of generating a classification structure.

Conceptual clustering is closely related to fuzzy set theory, in which objects may belong to one or more groups, in varying degrees of fitness.