Local Neural Networks of Space-Time Predicting Modeling for Lattice Data in GIS

  • Haiqi Wang
  • Jinfeng Wang
  • Xuhua Liu
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3973)


Lattice data have two different scale spatial properties: global depen- dence property and local fluctuation property. For lattice data space-time autoregressive modeling, to reduce influence of spatial fluctuation on prediction accuracy of neural networks, all regions are partitioned into several subareas by an improved k-means algorithm based on spatial contiguity relation. Some partition criteria are proposed to evaluate different partition schemes and the optimal scheme has the least spatial fluctuation and significant spatial dependent within each subarea. Each multi-layer perceptrons (MLPs) network is modeled respectively for each subarea, and the output nodes are the prediction values at time t of an attribute for all regions in a subarea, and the input nodes are observations before time t of this subarea itself and neighboring regions. As a case study, all local models are tested and compared with a single global MLPs network by one-step-ahead predicting of an epidemic dataset, and the results indicate that local NN model has better prediction performance than the global NN model.


Lattice Data Output Node Input Node Partition Scheme Spatial Fluctuation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Cressie, N.A.: Statistics for Spatial Data, 2nd edn. Wiley, New York (1993)Google Scholar
  2. 2.
    Tobler, W.: A Computer Movie Simulating Urban Growth in the Detroit Region. Economic Geography 46, 234–240 (1970)CrossRefGoogle Scholar
  3. 3.
    Haining, R., Wise, S., Ma, J.: Designing and Implementing Software for Spatial Statistical Analysis in a GIS Environment. Journal of Geographical Systems 2, 257–286 (2000)CrossRefGoogle Scholar
  4. 4.
    Anselin, L.: Spatial Econometrics: Methods and Models. Kluwer Academic Publishers, Dordrecht (1988)Google Scholar
  5. 5.
    Gilardi, N., Bengio, S.: Local Machine Learning Models for Spatial Data Analysis. Journal of Geographic Information and Decision Analysis 4(1), 11–28 (2000)Google Scholar
  6. 6.
    Anselin, L.: Local Indicators of Spatial Association–LISA. Geographical Analysis 27(2), 93–115 (1995)CrossRefGoogle Scholar
  7. 7.
    Wise, S., Haining, R., Ma, J.: Regionalization Tools for the Exploratory Spatial Analysis of Health Data. In: Fisher, M., Getis, A. (eds.) Recent Developments in Spatial Analysis: Spatial Statistics, Behavioral Modeling, and Computational Intelligence. Springer, Heidelberg (1997)Google Scholar
  8. 8.
    Hu, T.M., Sung, S.Y.: Data Fusion in Radial Basis Function Networks for Spatial Regression. Neural Processing Letters 21, 81–93 (2005)CrossRefGoogle Scholar
  9. 9.
    Anselin, L.: GeoDa 0.9 User’s Guide (2003),
  10. 10.
    Zhang, G., Patuwo, B.E., Hu, M.Y.: Forecasting with Artificial Neural Networks: the State of the Art. International Journal of Forecasting 14, 35–62 (1998)CrossRefGoogle Scholar
  11. 11.
    Fotheringham, A.S., O’Kelly, M.E.: Spatial Interaction Models: Formulations and Applications. Kluwer Academic Publishers, Dordrecht (1989)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Haiqi Wang
    • 1
    • 2
  • Jinfeng Wang
    • 1
  • Xuhua Liu
    • 3
  1. 1.Institute of Geographical Sciences and Nature Resources ResearchChinese Academy of SciencesBeijingP.R. China
  2. 2.College of Geo-resources and InformationUniversity of Petroleum (East China)DongyingP.R. China
  3. 3.Department of Electrical EngineeringUniversity of Southern CaliforniaLos AngelesU.S.A.

Personalised recommendations