Dealing with spatial autocorrelation when learning predictive clustering trees

Abstract

Spatial autocorrelation is the correlation among data values which is strictly due to the relative spatial proximity of the objects that the data refer to. Inappropriate treatment of data with spatial dependencies, where spatial autocorrelation is ignored, can obfuscate important insights. In this paper, we propose a data mining method that explicitly considers spatial autocorrelation in the values of the response (target) variable when learning predictive clustering models. The method is based on the concept of predictive clustering trees (PCTs), according to which hierarchies of clusters of similar data are identified and a predictive model is associated to each cluster. In particular, our approach is able to learn predictive models for both a continuous response (regression task) and a discrete response (classification task). We evaluate our approach on several real world problems of spatial regression and spatial classification. The consideration of the autocorrelation in the models improves predictions that are consistently clustered in space and that clusters try to preserve the spatial arrangement of the data, at the same time providing a multi-level insight into the spatial autocorrelation phenomenon. The evaluation of SCLUS in several ecological domains (e.g. predicting outcrossing rates within a conventional field due to the surrounding genetically modified fields, as well as predicting pollen dispersal rates from two lines of plants) confirms its capability of building spatial aware models which capture the spatial distribution of the target variable. In general, the maps obtained by using SCLUS do not require further post-smoothing of the results if we want to use them in practice.

Transductive hyperspectral image classification: toward integrating spectral and relational features via an iterative ensemble system

“…In any case, an emerging trend in hyperspectral imaging is to accommodate spatial correlation into the hyperspectral classification process as this improves predictive performance (Plaza et al 2009;Fauvel et al 2013). This improvement was also observed by recent machine learning studies in other predictive tasks involving spatial correlation such as regression, interpolation and forecasting (Stojanova et al 2013;). Further motivation is driven by the recently emerged perspective on the importance of a relational learning approach in spatial data mining (Malerba 2008).…”

Big Data Techniques For Supporting Accurate Predictions of Energy Production From Renewable Sources

“…From Table 1, we can also notice that geographic coordinates improve the prediction effectiveness, suggesting that data are subject to spatial autocorrelation phenomena. [34].…”

Spatial cluster detection using nearest neighbor distance

“…Since detecting spatial pattern of events is essential in many elds (medicine, cosmology with spatial clustering of galaxies, social sciences and criminology, agronomy and more), a substantial literature has been dedicated to the issue of spatial clustering (Murray et al, 2014). The two most popular cluster detection approaches are the spatial scan statistics (Kulldor and Nagarwalla, 1995;Kulldor, 1997;Patil and Taillie, 2004;Duczmal and Assuncao, 2004;Tango and Takahashi, 2005;Demattei et al, 2007) and spatial autocorrelation (Ord and Getis, 1995;F Dormann et al, 2007;Stojanova et al, 2013). On one hand, spatial scan statistics aim at scanning the studied area using windows of an imposed shape (circles, ellipses or squares): based on a likelihood ratio test, spatial clusters are dened by the windows that group together an abnormally high number of cases.…”