Identification of Clusters from Data Sets

Mufa, Vesna and Manevska, Violeta (2013) Identification of Clusters from Data Sets. 3rd International Conference on Application of Information and Communication Technology and Statistics in Economy and Education 2013, (ICAICTSEE-2013), UNWE, Sofia, Bulgaria, pp. 307-313, 200. pp. 307-313. ISSN ISBN 978-954-644-586-5

Full text not available from this repository.

Abstract

The era in which we live is called information age, i.e. the era of data. The evolution of data mining began with the storage of data in a computer. Data mining is analyzing of large data sets in order to find unexpected relationships and summaries of data on new, previously unknown, comprehensible and useful ways. One of the most common tasks of data mining is a descriptive modeling, which describes data and processes that they generate. One of the most used descriptive methods is а clustering. The clustering aims to detect natural groups (clusters) into the data. The purpose of this paper is the identification of clusters, their presentation and visualization. The clusters are identified through partitioning, as the basic version of clustering, which divides the instances of several mutually exclusive clusters, with using of the Euclidean distance. Determination of the exact number of clusters is important because the correct number of clusters is not only an input parameter, but it controls the granularity in cluster analysis too.

Item Type: Article
Subjects: Scientific Fields (Frascati) > Natural sciences > Computer and information sciences
Divisions: Faculty of Information and Communication Technologies
Depositing User: Mr Vladimir Milevski
Date Deposited: 24 Jun 2016 09:23
Last Modified: 24 Jun 2016 09:23
URI: https://eprints.uklo.edu.mk/id/eprint/37

Actions (login required)

View Item View Item