Building clusters for pattern recognition and analysis of geographical areas can be a useful way to provide relevant information for economic and social decisions. In this paper, we introduce a novel spatial clustering technique, called Bootstrap ClustGeo (BCG), which is a hierarchical approach, based on bootstrap techniques with spatial constraints. We evaluate the performance of the proposed approach BCG through some real case studies and simulations studies with different complexity, by computing Clustering Validation Measures (CVM) and then we compare the approach with the recently proposed ClustGeo (CG). These analyses exhibit the accuracy of BCG, also with respect to CG, in the presented applications, and highlight the great potentiality of this new clustering technique to provide meaningful information for spatial analysis. (C) 2020 Elsevier B.V. All rights reserved.
Identifying spatial patterns with the Bootstrap ClustGeo technique
Distefano V.;
2020-01-01
Abstract
Building clusters for pattern recognition and analysis of geographical areas can be a useful way to provide relevant information for economic and social decisions. In this paper, we introduce a novel spatial clustering technique, called Bootstrap ClustGeo (BCG), which is a hierarchical approach, based on bootstrap techniques with spatial constraints. We evaluate the performance of the proposed approach BCG through some real case studies and simulations studies with different complexity, by computing Clustering Validation Measures (CVM) and then we compare the approach with the recently proposed ClustGeo (CG). These analyses exhibit the accuracy of BCG, also with respect to CG, in the presented applications, and highlight the great potentiality of this new clustering technique to provide meaningful information for spatial analysis. (C) 2020 Elsevier B.V. All rights reserved.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
