Difference: Papers20100623104752 (2 vs. 3)

Revision 32010-08-13 - TheresiaFreska

 
META TOPICPARENT name="AI_GEOSTATSPapers"
Title: Rainfall Prediction using Artificial Neural Networks
Changed:
<
<
Date:
>
>
Date: 1 May 1998
  Authors: S. Lee, S. Cho and P. M. Wong
Changed:
<
<
Link: fileadmin/Documents/SIC97_GIDA/LeeSY.pdf
>
>
Link: LeeSY.pdf
  Abstract:
Added:
>
>
The spatial interpolation comparison 97 is concerned with predicting the daily rainfall at 367 locations based on the daily rainfall at nearby 100 locations in Switzerland. We propose a divide-and-conquer approach where the whole region is divided into four sub-areas and each is modeled with a different method. Predictions in two larger areas were made by RBF networks based on the locational information only. The two smaller areas were assumed to be implemented by the orographic effect which dictates that precipitation is proportional to elevation. Thus, predictions in these two areas were made using a simple linear regression model based on the elevation information only. Comparison with the observed data revealed that RBF networks produced good predictions while the linear models poor predictions. The relatively large prediction errors from the small areas seem to indicate that the orographic effect did not exist.
 REFERENCE

Journal of Geographic Information and Decision Analysis, Vol. 2. No. 2., pp. 233-242, 1998.

Changed:
<
<
Abstract
>
>
Keywords: Divide and conquer, neural network, orographic effect, radial basis function network, universal function approximation.
 
Changed:
<
<
The spatial interpolation comparison 97 is concerned with predicting the daily rainfall at 367 locations based on the daily rainfall at nearby 100 locations in Switzerland. We propose a divide-and-conquer approach where the whole region is divided into four sub-areas and each is modeled with a different method. Predictions in two larger areas were made by RBF networks based on the locational information only. The two smaller areas were assumed to be implemented by the orographic effect which dictates that precipitation is proportional to elevation. Thus, predictions in these two areas were made using a simple linear regression model based on the elevation information only. Comparison with the observed data revealed that RBF networks produced good predictions while the linear models poor predictions. The relatively large prediction errors from the small areas seem to indicate that the orographic effect did not exist.
>
>
META FILEATTACHMENT attachment="LeeSY.pdf" attr="h" comment="" date="1281731291" name="LeeSY.pdf" path="LeeSY.pdf" size="143161" stream="LeeSY.pdf" user="Main.TheresiaFreska" version="1"
Deleted:
<
<
Keywords:

Divide and conquer, neural network, orographic effect, radial basis function network, universal function approximation.

-- TWikiAdminUser - 2010-06-16

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