Impact areas in spatial analysis are determined by using density clustering methods which have an elevated computational complexity. We propose an al-ternative method based on the Extended Fuzzy C-Means (EFCM) method which has the following advantages: robustness to noise and outliers, linear computational complexity, automatic determination of the optimal number of clusters. In this method we derive dynamic buffer areas as hypersphere volume prototypes which become circles in the case of bi-dimensional pattern data: for example, event point data in Geographic Information Systems (GIS). In spatial analysis these circles can represent buffer areas around unknown epicenters (for example, crime events, desease localizations, car incidents). We have imple-mented the EFCM algorithm in a GIS created with the usage of ESRI/ARCGIS and ESRI/ARCVIEW 3.3 software tools and we have applied it to the analysis of hotspot fire events in the Santa Fè district (NM).
Dynamic buffer areas obtained by EFCM method in GIS environment / Sessa, Salvatore; DI MARTINO, F.. - STAMPA. - 5188:(2008), pp. 92-95. (Intervento presentato al convegno 10th International Conference VISUAL 2008 tenutosi a Salerno nel 11-12 settembre 2008) [10.1007/978-3-540-85891-1_11].
Dynamic buffer areas obtained by EFCM method in GIS environment
SESSA, SALVATORE;F. DI MARTINO
2008
Abstract
Impact areas in spatial analysis are determined by using density clustering methods which have an elevated computational complexity. We propose an al-ternative method based on the Extended Fuzzy C-Means (EFCM) method which has the following advantages: robustness to noise and outliers, linear computational complexity, automatic determination of the optimal number of clusters. In this method we derive dynamic buffer areas as hypersphere volume prototypes which become circles in the case of bi-dimensional pattern data: for example, event point data in Geographic Information Systems (GIS). In spatial analysis these circles can represent buffer areas around unknown epicenters (for example, crime events, desease localizations, car incidents). We have imple-mented the EFCM algorithm in a GIS created with the usage of ESRI/ARCGIS and ESRI/ARCVIEW 3.3 software tools and we have applied it to the analysis of hotspot fire events in the Santa Fè district (NM).I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
