A4 Article in conference proceedings
Comparison of cluster validation indices with missing data (2018)

Niemelä, M., Äyrämö, S., & Kärkkäinen, T. (2018). Comparison of cluster validation indices with missing data. In ESANN 2018 : Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (pp. 461-466). ESANN. https://www.elen.ucl.ac.be/Proceedings/esann/esannpdf/es2018-16.pdf

JYU authors or editors

Niemelä, Marko
Kärkkäinen, Tommi
Äyrämö, Sami

Publication details

All authors or editors: Niemelä, Marko; Äyrämö, Sami; Kärkkäinen, Tommi

Parent publication: ESANN 2018 : Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

Conference:

European symposium on artificial neural networks, computational intelligence and machine learning

ISBN: 978-2-87587-047-6

Publication year: 2018

Pages range: 461-466

Publisher: ESANN

Publication country: Belgium

Publication language: English

Persistent website address: https://www.elen.ucl.ac.be/Proceedings/esann/esannpdf/es2018-16.pdf

Publication open access: Other way freely accessible online

Publication channel open access:

Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/62745

Additional information: ESANN 2018 : 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Bruges, Belgium, April 25-26-27, 2018.

Abstract

Clustering is an unsupervised machine learning technique, which aims to divide a given set of data into subsets. The number of hidden groups in cluster analysis is not always obvious and, for this purpose, various cluster validation indices have been suggested. Recently some studies reviewing validation indices have been provided, but any experiments against missing data are not yet available. In this paper, performance of ten well-known indices on ten synthetic data sets with various ratios of missing values is measured using squared euclidean and city block distances based clustering. The original indices are modified for a city block distance in a novel way. Experiments illustrate the different degree of stability for the indices with respect to the missing data.

Keywords: data; cluster analysis

Free keywords: clustering; cluster validation

Fields of science: