Authors:
М.V. Nazarenko, Student of the State Higher Educational Institution “National Mining University”, Dnipropetrovsk, Ukraine
L.V. Sarycheva, Candidate of Sciences (Phys.-Math.), Associate Professor, Professor of the GIS Department of the State Higher Educational Institution “National Mining University”, Dnipropetrovsk, Ukraine
Annotation:
Authors propose a mathematical model and a method of clustering which takes into account the intuitive idea of grouping the data without imposing a priori assumptions about data structures. Clustering method FuzzyCluster was developed on the base of fuzzy description, capable of functioning under conditions of uncertainty about data structures, as well as that which takes into account the intuitive idea of data grouping. When comparing results of the five standard datasets clustering made by the algorithm FuzzyCluster with those made by the algorithms k-means and c-means we can see the advantages of the FuzzyCluster.
Bibliography:
1. Clustering based on fuzzy relations. Algorithm FuzzyRelationClustering. URL: http://www.spellabs.ru/download/FuzzyRelationClustering.doc
2. B. Duran and P. Odell. Cluster analysis. Translation. from English. E.Z. Demidenko. Ed. A.J. Bojarski. – M.: "Statistics", 1977. – 128 p.
3. Zagoruiko NG Applied methods of data analysis and knowledge. – Novosibirsk.: Institute of Mathematics, 1999 – 270.
4. Data mining using clustering. URL: http://www.kg.ru/?page_id=183
5. Algorithms for cluster analysis. URL: http://www.dea-analysis.ru/clustering-5.htm
6. Clustering algorithms in the service of Data Mining. URL: http://www.basegroup.ru/library/analysis/clusterization/datamining/
![]() ![]() |
|
![]() ![]() ![]() |
![]() |