Identifying single good clusters in data sets

Authors:
Frank Klawonn
Affiliations:
Department of Computer Science, University of Applied Sciences Braunschweig/Wolfenbuettel, Wolfenbuettel, Germany
Venue:
IWICPAS'06 Proceedings of the 2006 Advances in Machine Vision, Image Processing, and Pattern Analysis international conference on Intelligent Computing in Pattern Analysis/Synthesis
Year:
2006

Citing 3
Cited 1

Characterization and detection of noise in clustering

Pattern Recognition Letters
Fuzzy Models and Algorithms for Pattern Recognition and Image Processing

Fuzzy Models and Algorithms for Pattern Recognition and Image Processing
Detecting groups of anomalously similar objects in large data sets

IDA'05 Proceedings of the 6th international conference on Advances in Intelligent Data Analysis

Effective FCM noise clustering algorithms in medical images

Computers in Biology and Medicine

Quantified Score

Hi-index	0.00

Visualization

Abstract

Local patterns in the form of single clusters are of interest in various areas of data mining. However, since the intention of cluster analysis is a global partition of a data set into clusters, it is not suitable to identify single clusters in a large data set where the majority of the data can not be assigned to meaningful clusters. This paper presents a new objective function-based approach to identify a single good cluster in a data set making use of techniques known from prototype-based, noise and fuzzy clustering. The proposed method can either be applied in order to identify single clusters or to carry out a standard cluster analysis by finding clusters step by step and determining the number of clusters automatically in this way.