Pattern clustering, a fundamental tool in exploratory data analysis, involves unsupervised classification of data items into groups. This study presents an overview of pattern clustering methods from a statistical pattern recognition perspective. It identifies cross-cutting themes and highlights recent advances to offer useful guidance to clustering practitioners. The study addresses the clustering problem in many contexts and from multiple disciplines. Different assumptions and contexts make the transfer of generic methodologies occur slowly. The paper presents a taxonomy of techniques. Applications of clustering algorithms, including image segmentation, object recognition, and information retrieval, are described. This review provides a foundation for researchers and practitioners in this field.
Published in ACM Computing Surveys, this paper fits the journal's purpose of providing comprehensive overviews of topics within computer science. By presenting a taxonomy of clustering techniques and discussing recent advances, the article offers valuable insights to the broad community of clustering practitioners.