Hierarchical spherical clustering

Authors:
Vicenç Torra;Sadaaki Miyamoto
Affiliations:
Institut d' Investigació en Intel-ligència Artificial - CSIC Campus UAB s/n, E-08193 Bellaterra, Catalunya, Spain;Institute of Engineering Mechanics and Systems Univeristy of Tsukuba, Ibaraki 305-8573, Japan
Venue:
International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems
Year:
2002

Citing 4
Cited 2

Self-organizing maps

Self-organizing maps
A multi-stage system in compilation environments

Fuzzy Sets and Systems
Fuzzy clustering with weighting of data variables

International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems - special issue on measures and aggregation: formal aspects and applications to clustering and decision
Pattern Recognition with Fuzzy Objective Function Algorithms

Pattern Recognition with Fuzzy Objective Function Algorithms

Exploration of textual document archives using a fuzzy hierarchical clustering algorithm in the GAMBAL system

Information Processing and Management: an International Journal - Special issue: Cross-language information retrieval
Image clustering for the exploration of video sequences: Special Topic Section on Soft Approaches to Information Retrieval and Information Access on the Web

Journal of the American Society for Information Science and Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

This work introduces an alternative representation for large dimensional data sets. Instead of using 2D or 3D representations, data is located on the surface of a sphere. Together with this representation, a hierarchical clustering algorithm is defined to analyse and extract the structure of the data. The algorithm builds a hierarchical structure (a dendrogram) in such a way that different cuts of the structure lead to different partitions of the surface of the sphere. This can be seen as a set of concentric spheres, each one being of different granularity. Also, to obtain an initial assignment of the data on the surface of the sphere, a method based on Sammon's mapping has been developed.