New indices for cluster validity assessment

Kim, Minho; Ramakrishna, R. S.

doi:10.1016/j.patrec.2005.04.007

Cited by 227 publications

(121 citation statements)

References 13 publications

Supporting

Mentioning

119

Contrasting

Order By: Relevance

“…The clustering algorithm must allow an input parameter that sets the number of clusters the generated partition will have, also known as the k parameter due to the well known k-means algorithm. The agglomerative hierarchical algorithm (Jain and Dubes, 1988) and the k-means algorithm (Mirkin, 2005) are widely used for this purpose (Kim and Ramakrishna, 2005;Maulik and Bandyopadhyay, 2002;Milligan and Cooper, 1985;Pal and Biswas, 1997).…”

Section: Classical Methodology To Evaluate Cvismentioning

confidence: 99%

“…Many authors did not even consider the algorithm correctness assumption and they carried out the experiment without checking if the P nc partitions obtained were correct or were at least the best partitions the algorithms used could propose (Günter and Bunke, 2003;Kim and Ramakrishna, 2005;Maulik and Bandyopadhyay, 2002;Pal and Biswas, 1997). This type of work may be valid and the conclusions drawn from it may be correct, but we cannot be confident about the results obtained without a validation step.…”

Section: Related Workmentioning

confidence: 99%

“…Many authors have proposed different indices, called internal cluster validity indices (Halkidi et al, 2001;Kim and Ramakrishna, 2005;Maulik and Bandyopadhyay, 2002), to perform this validation. Unfortunately, no internal CVI has proved to be efficient in all conditions.…”

Section: Introductionmentioning

confidence: 99%

“…In addition, most new CVI proposals have been compared to well known indices (Chou et al, 2004;Hardy, 1996;Kim and Ramakrishna, 2005;Kothari and Pitts, 1999;Pal and Biswas, 1997). Therefore, it is clear that a method for comparing different CVIs, which we call an evaluation of cluster validity indices, is necessary.…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, it is clear that a method for comparing different CVIs, which we call an evaluation of cluster validity indices, is necessary. Although little theoretical work has been done in this context, most CVI evaluation work has followed the same methodology (Chou et al, 2004;Devillez et al, 2002;Günter and Bunke, 2003;Kim and Ramakrishna, 2005;Maulik and Bandyopadhyay, 2002;Milligan and Cooper, 1985;Pal and Biswas, 1997). We call this methodology the classical methodology.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

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: Classical Methodology To Evaluate Cvismentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

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

References

2010

A Statistical Approach to Genetic Epidemiology

View full text Add to dashboard Cite

Genome‐wide association studies: quality control and population‐based measures

Ziegler

2009

Genetic Epidemiology

View full text Add to dashboard Cite

Genome-wide association studies using hundreds of thousands of single-nucleotide polymorphism (SNP) markers have become a standard approach for identifying disease susceptibility genes. The change in the technology poses substantial computational and statistical challenges that have been addressed in the quality control, imputation, and population-based measure groups of the Genetic Analysis Workshop 16. The computational challenges pertain to efficient memory management and computational speed of the statistical procedures, and we discuss an approach for efficient SNP storage. Accuracy and computational speed is relevant for genotype calling, and the results from a comparison of three calling algorithms are discussed. The first statistical challenge is related to statistical quality control, and we discuss two novel quality control procedures. These low-level analyses have an effect on subsequent preparatory steps for high-level analyses, e.g., the quality of genotype imputation approaches. After the conduct of a genome-wide association study with successful replication and/or validation, measures of diagnostic accuracy including the area under the curve are investigated. The area under the curve can be constructed from summary data in some situations. Finally, we discuss how the population-attributable risk of a genetic variant that is only measured in a reference data set can be determined.

show abstract

New indices for cluster validity assessment

Cited by 227 publications

References 13 publications

Towards a standard methodology to evaluate internal cluster validity indices

Towards a standard methodology to evaluate internal cluster validity indices

References

Genome‐wide association studies: quality control and population‐based measures

Contact Info

Product

Resources

About