Number of the records: 1
K-Means Clustering for Problems with Periodic Attributes
- 1.
SYSNO ASEP 0328432 Document Type J - Journal Article R&D Document Type Journal Article Subsidiary J Článek ve WOS Title K-Means Clustering for Problems with Periodic Attributes Title Shlukovací algoritmus K-Means v problémech s periodickými atributy Author(s) Vejmelka, Martin (UIVT-O) SAI, RID, ORCID
Musílek, P. (CA)
Paluš, Milan (UIVT-O) RID, SAI, ORCID
Pelikán, Emil (UIVT-O) SAI, RIDSource Title International Journal of Pattern Recognition and Artificial Intelligence - ISSN 0218-0014
Roč. 23, č. 4 (2009), s. 721-743Number of pages 23 s. Language eng - English Country SG - Singapore Keywords clustering algorithms ; similarity measures ; K-means ; periodic attributes Subject RIV BB - Applied Statistics, Operational Research R&D Projects 1ET400300513 GA AV ČR - Academy of Sciences of the Czech Republic (AV ČR) CEZ AV0Z10300504 - UIVT-O (2005-2011) UT WOS 000267117500003 EID SCOPUS 67650703129 DOI 10.1142/S0218001409007338 Annotation The K-means algorithm is very popular in the machine learning community due to its inherent simplicity. However, in its basic form, it is not suitable for use in problems which contain periodic attributes, such as oscillator phase, hour of day or directional heading. A commonly used technique of trigonometrically encoding periodic input attributes to artificially generate the required topology introduces a systematic error. In this paper, a metric which induces a conceptually correct topology for periodic attributes is embedded into the K-means algorithm. This requires solving a non-convex minimization problem in the maximization step. Results of numerical experiments comparing the proposed algorithm to K-means with trigonometric encoding on synthetically generated data are reported. The advantage of using the proposed K-means algorithm is also shown on a real example using gas load data to build simple predictive models. Workplace Institute of Computer Science Contact Tereza Šírová, sirova@cs.cas.cz, Tel.: 266 053 800 Year of Publishing 2010
Number of the records: 1