Local Machine Learning Models for Spatial Data Analysis

Local Machine Learning Models for Spatial Data Analysis. Gilardi, N. & Bengio, S. Journal of Geographic Information and Decision Analysis, 4(1):11–28, 2000.

Paper abstract bibtex

In this paper, we compare different machine learning algorithms applied to non stationary spatial data analysis. We show that models taking into account local variability of the data are better than models which are trained globally on the whole dataset. Two global models (Support Vector Regression and Multilayer Perceptrons) and two local models (a local version of Support Vector Regression and Mixture of Experts) were compared over the Spatial Interpolation Comparison 97 (SIC97) dataset, and the results are presented and compared to previous results obtained on the same dataset.

@article{gilardi:2000:gida,
  author = {N. Gilardi and S. Bengio},
  title = {Local Machine Learning Models for Spatial Data Analysis},
  journal = {Journal of Geographic Information and Decision Analysis},
  volume = 4,
  number = 1,
  pages = {11--28},
  year = {2000},
  url = {publications/ps/rr00-34.ps.gz},
  pdf = {publications/pdf/rr00-34.pdf},
  djvu = {publications/djvu/rr00-34.djvu},
  original={2000/spatial_gida},
  web = {http://www.geodec.org/gida_7.htm},
  topics = {geostats},
  abstract = {In this paper, we compare different machine learning algorithms applied to non stationary spatial data analysis. We show that models taking into account local variability of the data are better than models which are trained globally on the whole dataset. Two global models (Support Vector Regression and Multilayer Perceptrons) and two local models (a local version of Support Vector Regression and Mixture of Experts) were compared over the Spatial Interpolation Comparison 97 (SIC97) dataset, and the results are presented and compared to previous results obtained on the same dataset.},
  categorie = {A}
}

Downloads: 0

{"_id":"ha4rzvvB5MQoSzHgP","bibbaseid":"gilardi-bengio-localmachinelearningmodelsforspatialdataanalysis-2000","authorIDs":[],"author_short":["Gilardi, N.","Bengio, S."],"bibdata":{"bibtype":"article","type":"article","author":[{"firstnames":["N."],"propositions":[],"lastnames":["Gilardi"],"suffixes":[]},{"firstnames":["S."],"propositions":[],"lastnames":["Bengio"],"suffixes":[]}],"title":"Local Machine Learning Models for Spatial Data Analysis","journal":"Journal of Geographic Information and Decision Analysis","volume":"4","number":"1","pages":"11–28","year":"2000","url":"publications/ps/rr00-34.ps.gz","pdf":"publications/pdf/rr00-34.pdf","djvu":"publications/djvu/rr00-34.djvu","original":"2000/spatial_gida","web":"http://www.geodec.org/gida_7.htm","topics":"geostats","abstract":"In this paper, we compare different machine learning algorithms applied to non stationary spatial data analysis. We show that models taking into account local variability of the data are better than models which are trained globally on the whole dataset. Two global models (Support Vector Regression and Multilayer Perceptrons) and two local models (a local version of Support Vector Regression and Mixture of Experts) were compared over the Spatial Interpolation Comparison 97 (SIC97) dataset, and the results are presented and compared to previous results obtained on the same dataset.","categorie":"A","bibtex":"@article{gilardi:2000:gida,\n author = {N. Gilardi and S. Bengio},\n title = {Local Machine Learning Models for Spatial Data Analysis},\n journal = {Journal of Geographic Information and Decision Analysis},\n volume = 4,\n number = 1,\n pages = {11--28},\n year = {2000},\n url = {publications/ps/rr00-34.ps.gz},\n pdf = {publications/pdf/rr00-34.pdf},\n djvu = {publications/djvu/rr00-34.djvu},\n original={2000/spatial_gida},\n web = {http://www.geodec.org/gida_7.htm},\n topics = {geostats},\n abstract = {In this paper, we compare different machine learning algorithms applied to non stationary spatial data analysis. We show that models taking into account local variability of the data are better than models which are trained globally on the whole dataset. Two global models (Support Vector Regression and Multilayer Perceptrons) and two local models (a local version of Support Vector Regression and Mixture of Experts) were compared over the Spatial Interpolation Comparison 97 (SIC97) dataset, and the results are presented and compared to previous results obtained on the same dataset.},\n categorie = {A}\n}\n\n","author_short":["Gilardi, N.","Bengio, S."],"key":"gilardi:2000:gida","id":"gilardi:2000:gida","bibbaseid":"gilardi-bengio-localmachinelearningmodelsforspatialdataanalysis-2000","role":"author","urls":{"Paper":"http://bengio.abracadoudou.com/publications/ps/rr00-34.ps.gz"},"downloads":0},"bibtype":"article","biburl":"http://bengio.abracadoudou.com/samy.bib","creationDate":"2020-03-18T03:43:27.315Z","downloads":0,"keywords":[],"search_terms":["local","machine","learning","models","spatial","data","analysis","gilardi","bengio"],"title":"Local Machine Learning Models for Spatial Data Analysis","year":2000,"dataSources":["9NCW2CDr4M3s5DvNX"]}