June 2000 Submitted to ICONIP 2000 ENVIRONMENTAL DATA MAPPING WITH SUPPORT VECTOR REGRESSION AND GEOSTATISTICS Mikhail
نویسندگان
چکیده
The paper presents decision-oriented mapping of pollution using hybrid models based on statistical learning theory (support vector regression or SVR) and spatial statistics (geostatistics). Adaptive and robust SVR approach is used to model non-linear large scale trends in the region and geostatistical models – spatial predictions and spatial simulations – are used to prepare decisionoriented maps: prediction maps along with maps of error variance and equiprobable digital models of the pollution based on conditional stochastic simulations. The quality of the proposed approach is tested with the validation data set not used for the model development. Real data on soil contamination by Chernobyl radionuclides in Russia is used as a case study.
منابع مشابه
Environmental Data Mapping with Support Vector Regression and Geostatistics
The paper presents decision-oriented mapping of pollution using hybrid models based on statistical learning theory (support vector regression or SVR) and spatial statistics (geostatistics). Adaptive and robust SVR approach is used to model non-linear large scale trends in the region and geostatistical models – spatial predictions and spatial simulations – are used to prepare decisionoriented ma...
متن کاملEnvironmental data mining and modeling based on machine learning algorithms and geostatistics
The paper presents some contemporary approaches to the spatial environmental data analysis, processing and presentation. The main topics are concentrated on the decision–oriented problems of environmental and pollution spatial data mining and modelling: valorisation and representativity of data with the help of exploratory data analysis, topological, statistical and fractal measures of monitori...
متن کاملMachine Learning for Automatic Environmental Mapping: When and How
This paper discusses the opportunity of using Machine Learning techniques in an automatic environmental mapping context, as was the case for the SIC2004 exercise. First, the Machine Learning methodology is quickly described and compared to Geostatistics. From there, some clues about when to apply Machine Learning are proposed, and what outcomes can be expected from this choice. Finally, three w...
متن کاملMetal accumulation in mosses across national boundaries: uncovering and ranking causes of spatial variation.
This study aimed at cross-border mapping metal loads in mosses in eight European countries in 1990, 1995, and 2000 and at investigating confounding factors. Geostatistics was used for mapping, indicating high local variances but clear spatial autocorrelations. Inference statistics identified differences of metal concentrations in mosses on both sides of the national borders. However, geostatist...
متن کاملSpectroscopic Based Quantitative Mapping of Contaminant Elements in Dumped Soils of a Copper Mine
Possibility of mapping the distribution of Arsenic and Chromium in a mining area was investigated using combination of (VNIR) reflectance spectroscopy and geostatistical analysis. Fifty five soil samples were gathered from a waste dump at Sarcheshmeh copper mine and VNIR reflectance spectra were measured in a laboratory. Savitzky- Golay first derivative was used as the main pre-processing metho...
متن کامل