Extended fuzzy C-means clustering algorithm for hotspot events in spatial analysis

Article type: Research Article

Authors: Di Martino, Ferdinando^{a; b} | Loia, Vincenzo^b | Sessa, Salvatore^{a; *}

Affiliations: [a] Università degli Studi di Napoli Federico II, Dipartimento di Costruzioni e Metodi Matematici in Architettura, Via Monteoliveto 3, 80134 Napoli, Italy | [b] Università degli Studi di Salerno, Dipartimento di Matematica e Informatica, Via Ponte Don Melillo, 84084 Fisciano (Salerno), Italy

Correspondence: [*] Corresponding author: S. Sessa, Tel.: +390812538907; Fax: +390812538909; E-mail: sessa@unina.it

Abstract: A known approach for the detection of hotspots is to use a cluster technique, which is an effective method for determining areas with elevated concentrations of localized events. We show that the extended fuzzy C-means (EFCM) algorithm works better than the classical FCM algorithm: indeed it determines automatically the initial number of clusters, it prevents the problem of shifting the clusters with low density area of data points in areas with higher density of such points and it finds the cluster volume prototypes as hyperspheres, here used for identifying hotspot areas in spatial analysis where the data are events geo-referenced as points on the geographic map. We have implemented the EFCM algorithm in a geographic information system (GIS) created with the usage of ESRI/ARCGIS and ESRI/ARCVIEW software tools and we have applied it to a specific problem of buildings maintenance.

Keywords: Extended fuzzy C-means, GIS, hotspot, spatial analysis

DOI: 10.3233/HIS-2008-5103

Journal: International Journal of Hybrid Intelligent Systems, vol. 5, no. 1, pp. 31-44, 2008