Title: Extending the SACOC algorithm through the Nyström method for dense manifold data analysis

Authors: Héctor D. Menéndez; Fernando E.B. Otero; David Camacho

Addresses: Department of Computer Science, University College London, Gower Street, London WC1E 6BT, UK ' School of Computing, University of Kent, Medway Building Chatham Maritime, Kent ME4 4AG, UK ' Department of Computer Science, Universidad Autónoma de Madrid, Tomas y Valiente 11, Cantoblanco, Madrid, Spain

Abstract: The growing amount of data demands new analytical methodologies to extract relevant knowledge. Clustering is one of the most competitive techniques in this context. Using a dataset as a starting point, clustering techniques blindly group the data by similarity. Among the different areas, manifold identification is currently gaining importance. Spectral-based methods, which are one of the main used methodologies, are sensitive to metric parameters and noise. In order to solve these problems, new bio-inspired techniques have been combined with different heuristics to perform the cluster selection, in particular for dense datasets, featured by areas of higher density. This paper extends a previous algorithm named spectral-based ant colony optimisation clustering (SACOC), used for manifold identification. We focus on improving it through the Nyström extension for dealing with dense data problems. We evaluated the new approach, called SACON, comparing it against online clustering algorithms and the Nyström extension of spectral clustering.

Keywords: ant colony optimisation; ACO; clustering; data mining; machine learning; spectral; Nyström; SACON; SACOC.

DOI: 10.1504/IJBIC.2017.10004321

International Journal of Bio-Inspired Computation, 2017 Vol.10 No.2, pp.127 - 135

Received: 05 Dec 2014
Accepted: 25 Nov 2015

Published online: 28 Jul 2017 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article