Spatial prediction of soil properties in temperate mountain regions using support vector regression

被引:68
作者
Ballabio, Cristiano [1 ]
机构
[1] Univ Milano Bicocca, Environm & Land Sci Dept, Milan, Italy
关键词
Digital soil mapping; Mountain regions; Support vector regression; Model comparison; CLASSIFICATION; LANDSLIDE; SELECTION;
D O I
10.1016/j.geoderma.2009.04.022
中图分类号
S15 [土壤学];
学科分类号
0903 ; 090301 ;
摘要
Digital soil mapping in mountain areas faces two major limitations: the small number of available observations and the non-linearity of the relations between environmental variables and soil properties. A possible approach to deal with these limitations involves the use of non-parametric models to interpolate soil properties of interest. Among the different approaches currently available, Support Vector Regression (SVR) seems to have several advantages over other techniques. SVR is a set of techniques in which model complexity is limited by the learning algorithm itself, which prevents overfitting. Moreover, the non-linear approximation of SVR is based on a kernel transformation of the data, which avoids the use of complex functions and is computationally feasible; while the resulting projection in feature space is especially suited for sparse datasets. A brief introduction to this methodology, a comparison with other popular methodologies and a framework for the application of this approach to a study site in the Italian Alps is discussed. (C) 2009 Elsevier B.V. All rights reserved.
引用
收藏
页码:338 / 350
页数:13
相关论文
共 49 条
  • [1] Acevedo J, 2007, LECT NOTES COMPUT SC, V4507, P284
  • [2] AIZERMAN MA, 1965, AUTOMAT REM CONTR+, V25, P821
  • [3] Large sackung along major tectonic features in the Central Italian Alps
    Ambrosi, C
    Crosta, GB
    [J]. ENGINEERING GEOLOGY, 2006, 83 (1-3) : 183 - 200
  • [4] [Anonymous], 1961, Adaptive Control Processes: a Guided Tour, DOI DOI 10.1515/9781400874668
  • [5] [Anonymous], Journal of machine learning research
  • [6] [Anonymous], 1993, C4.5: Programs for machine learning
  • [7] [Anonymous], 1964, AUTOMATION REMOTE CO
  • [8] Barakat N., 2005, International Journal of Computational Intelligence, V2, P59
  • [9] Barakat N., 2004, 14 INT C COMP THEOR
  • [10] Soil hydraulic behaviour of a selected benchmark soil involved in the landslide of Sarno 1998
    Basile, A
    Mele, G
    Terribile, F
    [J]. GEODERMA, 2003, 117 (3-4) : 331 - 346