Modifying the Scale-free Clustering Method

Author

Päivinen, N.S. ; Grönfors, T.K.

Author_Institution

Dept. of Comput. Sci., Kuopio Univ.

Volume

2

fYear

2005

fDate

28-30 Nov. 2005

Firstpage

477

Lastpage

483

Abstract

The aim of this study is to computationally classify, without supervision, the data points of a dataset containing real-life measurements pre-classified into two classes. The classification is done using two methods: the k-means method as a reference, and a modified version of previously presented method using a minimum spanning tree with a scale-free structure. The construction of a scale-free minimum spanning tree (SFMST) and its usage in clustering are presented, and the results of the modified SFMST clustering method are compared with the results obtained using the k-means method

Keywords

pattern classification; pattern clustering; tree data structures; trees (mathematics); unsupervised learning; dataset; k-means method; minimum spanning tree; pattern classification; scale-free clustering method; Classification tree analysis; Clustering algorithms; Clustering methods; Complex networks; Computer science; Graph theory; Measurement errors; Network topology; Spatial databases; Tree graphs;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence for Modelling, Control and Automation, 2005 and International Conference on Intelligent Agents, Web Technologies and Internet Commerce, International Conference on

Conference_Location

Vienna

Print_ISBN

0-7695-2504-0

Type

conf

DOI

10.1109/CIMCA.2005.1631514

Filename

1631514