15
views
0
recommends
+1 Recommend
1 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      SELEÇÃO DE DENSIDADE DE AMOSTRAGEM COM BASE EM DADOS DE ÁREAS JÁ MAPEADAS PARA TREINAMENTO DE MODELOS DE ÁRVORE DE DECISÃO NO MAPEAMENTO DIGITAL DE SOLOS Translated title: SELECTION OF SAMPLING DENSITY BASED ON DATA FROM AREAS ALREADY MAPPED FOR TRAINING DECISION TREE MODELS IN DIGITAL SOIL MAPPING

      research-article

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Para estudar técnicas de amostragem, úteis ao mapeamento digital de solos (MDS), objetivou-se avaliar o efeito da variação da densidade de pontos amostrais com base em dados de áreas já mapeadas por métodos tradicionais na acurácia dos modelos de árvores de decisão (AD) para a geração de mapas de solos por MDS. Em duas bacias hidrográficas no noroeste do Rio Grande do Sul, usou-se, como referência, antigos mapas convencionais de solos na escala 1:50.000. A partir do modelo digital de elevação do terreno e da rede hidrográfica, foram gerados mapas das variáveis preditoras: elevação, declividade, curvatura, comprimento de fluxo, acúmulo de fluxo, índice de umidade topográfica e distância euclideana de rios. A escolha dos locais dos pontos amostrais foi aleatória e testaram-se densidades amostrais que variaram de 0,1 a 4 pontos/ha. O treinamento dos modelos foi realizado no software Weka, gerando-se modelos preditores usando diferentes tamanhos do nó final da AD para obter AD com tamanhos distintos. Quando não se controlou o tamanho das AD, o aumento da densidade de amostragem resultou no aumento da concordância com os mapas básicos de referências e no aumento do número de unidades de mapeamento preditas. Nas AD com tamanho controlado, o aumento da densidade de amostragem não influenciou a concordância com os mapas de referência e interferiu muito pouco no número de unidades de mapeamento preditas.

          Translated abstract

          In order to study sampling techniques useful for digital soil mapping (DSM), we evaluated the effect of changes in sampling density, based on data from areas already mapped by traditional methods, in regard to the accuracy of decision trees models for generating soil maps using DSM. In two watersheds in northwestern Rio Grande do Sul, Brazil, 1:50,000 scale conventional soils maps were used as reference maps. From the ASTER - GDEM Global Digital Elevation Model and the hydrographic network, maps of predictive variables were generated: elevation, slope, curvature, flow length, flow accumulation, topographic wetness index, and Euclidian distance of the streams. We used random sampling, and tested sampling densities that ranged from 0.1 to 4 points per hectare. Models were trained using Weka software, generating predictive models using different sizes of the final node to obtain decision trees of different sizes. The results indicate that when the size of the decision tree was not controlled, an increase in sampling density resulted in greater overall accuracy in accordance with the basic reference maps and an increase in the number of predicted soil mapping units. When the size of decision trees was controlled, an increase in sampling density did not affect the overall accuracy and had a very slight influence on the number of predicted mapping units.

          Related collections

          Most cited references20

          • Record: found
          • Abstract: not found
          • Article: not found

          Spatial data mining for enhanced soil map modelling

            Bookmark
            • Record: found
            • Abstract: not found
            • Article: not found

            Mapping soil landscape as spatial continua: The Neural Network Approach

            A-Xing Zhu (2000)
              Bookmark
              • Record: found
              • Abstract: found
              • Article: found
              Is Open Access

              Decision trees for digital soil mapping on subtropical basaltic steeplands

              When soil surveys are not available for land use planning activities, digital soil mapping techniques can be of assistance. Soil surveyors can process spatial information faster, to assist in the execution of traditional soil survey or predict the occurrence of soil classes across landscapes. Decision tree techniques were evaluated as tools for predicting the ocurrence of soil classes in basaltic steeplands in South Brazil. Several combinations of types of decicion tree algorithms and number of elements on terminal nodes of trees were compared using soil maps with both original and simplified legends. In general, decision tree analysis was useful for predicting occurrence of soil mapping units. Decision trees with fewer elements on terminal nodes yield higher accuracies, and legend simplification (aggregation) reduced the precision of predictions. Algorithm J48 had better performance than BF Tree, RepTree, Random Tree, and Simple Chart.
                Bookmark

                Author and article information

                Contributors
                Role: ND
                Role: ND
                Role: ND
                Journal
                rbcs
                Revista Brasileira de Ciência do Solo
                Rev. Bras. Ciênc. Solo
                Sociedade Brasileira de Ciência do Solo (Viçosa )
                1806-9657
                August 2015
                : 39
                : 4
                : 960-967
                Affiliations
                [1 ] Universidade Federal do Rio Grande do Sul Brazil
                [2 ] Universidade Federal do Rio Grande do Sul Brazil
                Article
                S0100-06832015000400960
                10.1590/01000683rbcs20140289
                d54b6ae8-d6d9-4e81-8690-a3f5128af40b

                http://creativecommons.org/licenses/by/4.0/

                History
                Product

                SciELO Brazil

                Self URI (journal page): http://www.scielo.br/scielo.php?script=sci_serial&pid=0100-0683&lng=en
                Categories
                SOIL SCIENCE

                Soil
                soil map,GIS,model,prediction,mapa de solos,SIG,modelo,predição
                Soil
                soil map, GIS, model, prediction, mapa de solos, SIG, modelo, predição

                Comments

                Comment on this article

                scite_

                Similar content183

                Cited by3

                Most referenced authors159