Difference: Papers20100623115526 (1 vs. 2)

Revision 22010-07-23 - TWikiAdminUser

Changed:
<
<
META TOPICPARENT name="GeostatisticsPapers"
>
>
META TOPICPARENT name="AI_GEOSTATSPapers"
 Title: Spatial Prediction of Radioactivity using General Regression Neural Network

Date:

Authors: Vadim Timonin, Elena Savelieva

Link: http://publications.epress.monash.edu/doi/pdf/10.2104/ag050019

Abstract:

This work describes an application of General Regression Neural Network (GRNN) to spatial predictions of radioactivity. GRNN belongs to a class of neural networks widely used for mapping continuous functions. It is based on a non-parametric (kernel) Parzen-Rosenblatt density estimator. The kernel size is the only tuning parameter, and it allows the user to implement a GRNN in an automatic mode. An important advantage of the GRNN is its very simple and fast training procedure. The most important drawbacks are high smoothing and dependence on the spatial density of the monitoring data set. The current case study is performed on the SIC2004 data sets, and the results obtained here can be compared with those obtained by the other participants using other approaches.

Reference

Applied GIS
Volume 1, No. 2, August 2005
DOI: 10.2104/ag050019

-- TWikiAdminUser - 2010-06-16

Revision 12010-06-23 - TWikiAdminUser

 
META TOPICPARENT name="GeostatisticsPapers"
Title: Spatial Prediction of Radioactivity using General Regression Neural Network

Date:

Authors: Vadim Timonin, Elena Savelieva

Link: http://publications.epress.monash.edu/doi/pdf/10.2104/ag050019

Abstract:

This work describes an application of General Regression Neural Network (GRNN) to spatial predictions of radioactivity. GRNN belongs to a class of neural networks widely used for mapping continuous functions. It is based on a non-parametric (kernel) Parzen-Rosenblatt density estimator. The kernel size is the only tuning parameter, and it allows the user to implement a GRNN in an automatic mode. An important advantage of the GRNN is its very simple and fast training procedure. The most important drawbacks are high smoothing and dependence on the spatial density of the monitoring data set. The current case study is performed on the SIC2004 data sets, and the results obtained here can be compared with those obtained by the other participants using other approaches.

Reference

Applied GIS
Volume 1, No. 2, August 2005
DOI: 10.2104/ag050019

-- TWikiAdminUser - 2010-06-16

 
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback