No abstract

No abstract

No abstract

In this article, multilayer perceptron (MLP) network models with spatial constraints are proposed for regionalization of geostatistical point data based on multivariate homogeneity measures. The study focuses on non‐stationarity and autocorrelation in spatial data. Supervised MLP machine learning algorithms with spatial constraints have been implemented and tested on a point dataset. MLP spatially weighted classification models and an MLP contiguity‐constrained classification model are developed to conduct spatially constrained regionalization. The proposed methods have been tested with an attribute‐rich point dataset of geological surveys in Ukraine. The experiments show that consideration of the spatial effects, such as the use of spatial attributes and their respective whitening, improve the output of regionalization. It is also shown that spatial sorting used to preserve spatial contiguity leads to improved regionalization performance.

Abstract. National and regional geographic datasets with derivative spatial analysis techniques are often crucial for decision-making in sustainable development for both developing and developed countries. The needs in the spatial data and respective training are demonstrated in this article by the two very diverse examples of Papua New Guinea and the Republic of Lithuania national spatial data infrastructures. Handling and efficient use of geographic data requires rather specific skills. However, the existing curricula are inconsistent and hardly match even the basic needs of geographic information managers in state institutions and municipalities. The main objective of the proposed geographic information e-training system is to develop and provide the modularised Spatial Information Infrastructure courses intended for on-line based learning. This mainly will target employees of civil service and private business in Lithuania and the European Union. The proposed curriculum is a set of modular courses adding up to 1,5-year part-time studies in the field of Geographic Information Science and Geographic Information Infrastructure. Main advantage of on-line-based training is increasing accessibility in terms of both geographic location and time.