Spatial Clustering Overview and Comparison: Accuracy, Sensitivity, and Computational Expense

Abstract

Cluster analysis continues to be an important exploratory technique in scientific inquiry. It is used widely in geography, public health, criminology, ecology, and many other fields. Spatial cluster detection is driven by geographic information corresponding to the location of activities, requiring appropriate and meaningful treatment of space and spatial relationships combined with observed attributes of location and events. To date, this has meant utilizing dedicated measures and techniques to structure and account for distance, neighbors, contiguity, irregular geographic morphology, and so on. Unfortunately, all spatial clustering approaches, regardless of their theoretical underpinning, statistical foundation, or mathematical specification, have limitations in accuracy, sensitivity, and the computational effort required for identifying clusters. As a result, a major challenge in practice is determining which technique(s) will provide the most meaningful insights for a particular substantive issue or planning context. The purpose of this article is to provide an overview and evaluation of spatial clustering techniques, identifying the strengths and weaknesses of the most widely applied approaches. Results suggest that performance varies significantly in terms of accuracy, sensitivity, and computational expense. This is noteworthy because the misidentification of clusters, whether false positives or false negatives, has the potential to bias not only hypothesis formulation but also pragmatic facets of policy, process, and planning efforts within a region. Key Words: cluster analysis, hot spots, knowledge discovery, method selection, scale.El an alisis de conglomerados [an alisis de clusters] sigue siendo una importante t ecnica exploratoria en investigaci on cient ıfica. Se le emplea ampliamente en geograf ıa, salud p ublica, criminolog ıa, ecolog ıa y en muchos otros campos. La detecci on de la aglomeraci on espacial est a controlada por la informaci on geogr afica correspondiente a la localizaci on de las actividades, lo cual requiere un tratamiento apropiado y significativo del espacio y de las relaciones espaciales combinadas con atributos observados de localizaci on y eventos. Hasta el momento presente, esto implica la utilizaci on de mediciones y t ecnicas especiales para estructurar y tomar en cuenta cosas como distancia, vecinos, contig€ uidad, morfolog ıa geogr afica irregular y dem as. Infortunadamente, todos los enfoques sobre aglomeraciones espaciales, sin consideraci on a sus bases te oricas, fundamentaci on estad ıstica, o especificaci on matem atica, adolecen de limitaciones en exactitud, sensibilidad y en el esfuerzo computacional que se necesita para identificar los conglomerados. Como resultado, en la pr actica un reto mayor es determinar qu e t ecnica o t ecnicas rendir an m as para un asunto sustantivo particular o un contexto de planificaci on. El prop osito de este art ıculo es proporcionar una mirada de conjunto y evaluaci on de t ecnicas de aglomeraci on espacial, identificando las forta...

An Adaptive Density-Based Time Series Clustering Algorithm: A Case Study on Rainfall Patterns

“…The literature provides several indexes that are useful for evaluating clustering results [17,18]. These indexes can evaluate the results of different clustering algorithms for the same dataset.…”

Çok Deği̇şkenli̇ Hari̇talama İçi̇n Kümeleme Yöntemleri̇ni̇n Kullanilmasi

“…[20] ve Grubesic ve ark. [21] yayınlarında seçilen kümeleme yöntemleriyle mekânsal analizler yapılmış ve kümeleme yöntemlerinin oluşturulan yapay verilerle hesaplama güçlüğü ve kümeleme başarısı açısından karşılaştırılması yapılmıştır. Guo ve ark.…”

Using K-Means and K-Medoids Methods for Multivariate Mapping

“…Different from the methods above, in order to represent many attributes in the same map, classification method based on clustering methods in data mining can be used as well. With the use of clustering methods, similar aspects of different spatial objects can be revealed by considering more than one attribute [18,19]. In this sense, spatial analyses that would make important contributions for risk analysis, planning etc.…”