A new cluster validity measure and its application to image compression

Chou, Chien-Hsing; Su, Mu-Chun; Lai, Eugene

doi:10.1007/s10044-004-0218-1

Cited by 258 publications

(174 citation statements)

References 38 publications

Supporting

Mentioning

171

Contrasting

Unclassified

Order By: Relevance

“…The goal of a clustering algorithm is to perform a partition where objects within a group are similar and objects in different groups are dissimilar. Therefore, the purpose of clustering is to identify natural structures in a dataset (Jain and Dubes, 1988;Halkidi et al, 2001;Mirkin, 2005;Sneath and Sokal, 1973) and it is widely used in many fields such as psychology (Holzinger and Harman, 1941), biology (Sneath and Sokal, 1973), pattern recognition (Mirkin, 2005), image processing (Chou et al, 2004) and computer security (Barbará and Jajodia, 2002).…”

Section: Introductionmentioning

confidence: 99%

Towards a standard methodology to evaluate internal cluster validity indices

Gurrutxaga

Muguerza

Arbelaitz

et al. 2011

Pattern Recognition Letters

View full text Add to dashboard Cite

The evaluation and comparison of internal cluster validity indices is a critical problem in the clustering area. The methodology used in most of the evaluations assumes that the clustering algorithms work correctly. We propose an alternative methodology that does not make this often false assumption. We compared 7 internal cluster validity indices with both methodologies and concluded that the results obtained with the proposed methodology are more representative of the actual capabilities of the compared indices.

show abstract

Section: Introductionmentioning

confidence: 99%

Towards a standard methodology to evaluate internal cluster validity indices

Gurrutxaga

Muguerza

Arbelaitz

et al. 2011

Pattern Recognition Letters

View full text Add to dashboard Cite

show abstract

“…The following real-life data sets are used in this paper which are taken from [2] [12]. Here, n is the number of data points, d is the number of features, and K is the number of clusters.…”

Section: Datasets Usedmentioning

confidence: 99%

“…In order to overcome the above-said issue, this paper proposes a clustering algorithm, where a number of trial solutions are provided with different cluster numbers along with cluster center coordinates for the same data set. Correction of each possible grouping is quantitatively evaluated with a global validity index, the CS measure [2]. Then, through the evolution mechanism, eventually, the best solutions start dominating the population, whereas the bad ones are eliminated.…”

Section: Introductionmentioning

confidence: 99%

Automatic Clustering Using Teaching Learning Based Optimization

Murty¹,

Naik²,

Murthy³

et al. 2014

View full text Add to dashboard Cite

Finding the optimal number of clusters has remained to be a challenging problem in data mining research community. Several approaches have been suggested which include evolutionary computation techniques like genetic algorithm, particle swarm optimization, differential evolution etc. for addressing this issue. Many variants of the hybridization of these approaches also have been tried by researchers. However, the number of optimal clusters and the computational efficiency has still remained open for further research. In this paper, a new optimization technique known as "Teaching-Learning-Based Optimization" (TLBO) is implemented for automatic clustering of large unlabeled data sets. In contrast to most of the existing clustering techniques, the proposed algorithm requires no prior knowledge of the data to be classified rather it determines the optimal number of partitions of the data "on the run". The new AUTO-TLBO algorithms are evaluated on benchmark datasets (collected from UCI machine repository) and performance comparisons are made with some well-known clustering algorithms. Results show that AUTO-TLBO clustering techniques have much potential in terms of comparative results and time of computations. KeywordsClustering, Auto-Clustering, TLBO M. R. Murty et al.

show abstract

“…For crisp clustering, some of the well-known indices available in the literature are the Dunn's index (DI) (Hertz et al, 2006;Dunn, 1974), Calinski-Harabasz index (Calinski and Harabasz, 1974), Davis-Bouldin (DB) index (Davies and Bouldin, 1979), PBM index (Pakhira et al, 2004), and the CS measure (Chou et al, 2004). In this work, we have based our fitness function on the CS measure as according to the authors, CS measure is more efficient in tackling clusters of different densities and/or sizes than the other popular validity measures, the price being paid in terms of high computational load with increasing k and n (Chou et al, 2004). Before applying the CS measure, centroid of a cluster is computed by averaging the data vectors belonging to that cluster using the formula,…”

Section: Reformulation Of Cs Measurementioning

confidence: 99%

“…Our experiments indicate that the proposed MEPSO algorithm yields more accurate results at a faster pace than the classical PSO in context to the present problem. (iii) We reformulate a recently proposed cluster validity index (known as the CS measure) (Chou et al, 2004) using the kernelized distance metric. The new CS measure forms the objective function to be minimized for optimal clustering.…”

Section: Introductionmentioning

confidence: 99%

Automatic kernel clustering with a Multi-Elitist Particle Swarm Optimization Algorithm

Das

Abraham

Konar

2008

Pattern Recognition Letters

170

View full text Add to dashboard Cite

This article introduces a scheme for clustering complex and linearly non-separable datasets, without any prior knowledge of the number of naturally occurring groups in the data. The proposed method is based on a modified version of classical Particle Swarm Optimization (PSO) algorithm, known as the Multi-Elitist PSO (MEPSO) model. It also employs a kernel-induced similarity measure instead of the conventional sum-of-squares distance. Use of the kernel function makes it possible to cluster data that is linearly non-separable in the original input space into homogeneous groups in a transformed high-dimensional feature space. A new particle representation scheme has been adopted for selecting the optimal number of clusters from several possible choices. The performance of the proposed method has been extensively compared with a few state of the art clustering techniques over a test suit of several artificial and real life datasets. Based on the computer simulations, some empirical guidelines have been provided for selecting the suitable parameters of the PSO algorithm.

show abstract

A new cluster validity measure and its application to image compression

Cited by 258 publications

References 38 publications

Towards a standard methodology to evaluate internal cluster validity indices

Towards a standard methodology to evaluate internal cluster validity indices

Automatic Clustering Using Teaching Learning Based Optimization

Automatic kernel clustering with a Multi-Elitist Particle Swarm Optimization Algorithm

Contact Info

Product

Resources

About