Data fusion in radial basis function networks for spatial regression

被引:12
作者
Hu, TM [1 ]
Sung, SY [1 ]
机构
[1] Natl Univ Singapore, Dept Comp Sci, Singapore 117543, Singapore
关键词
data fusion; spatial autocorrelation; spatial regression; radial basis function network;
D O I
10.1007/s11063-004-7776-5
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Conventional radial basis function (RBF) networks for spatial regression assume independent and identical distribution and ignore spatial information. In contrast to input fusion, we push spatial information further into RBF networks by fusing output from hidden and output layers. Three case studies demonstrate the advantage of hidden fusion over others and indicate the optimal value is around 1 for the coefficient used in hidden fusion, which links the output from the hidden layer for each site with their neighbors.
引用
收藏
页码:81 / 93
页数:13
相关论文
共 27 条
[1]  
Allen R. G., 1998, FAO Irrigation and Drainage Paper
[2]   Convergence of an EM-type algorithm for spatial clustering [J].
Ambroise, C ;
Govaert, G .
PATTERN RECOGNITION LETTERS, 1998, 19 (10) :919-927
[3]  
Bishop C. M., 1996, Neural networks for pattern recognition
[4]  
Cressie N, 1993, STAT SPATIAL DATA
[5]   Combining RBF networks trained by different clustering techniques [J].
de Carvalho, A ;
Brizzotti, MM .
NEURAL PROCESSING LETTERS, 2001, 14 (03) :227-240
[6]   Spatial data mining: A database approach [J].
Ester, M ;
Kriegel, HP ;
Sander, J .
ADVANCES IN SPATIAL DATABASES, 1997, 1262 :47-66
[7]  
Friedman J., 2001, ELEMENTS STAT LEARNI, V1
[8]   STOCHASTIC RELAXATION, GIBBS DISTRIBUTIONS, AND THE BAYESIAN RESTORATION OF IMAGES [J].
GEMAN, S ;
GEMAN, D .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1984, 6 (06) :721-741
[9]  
GILARDI N, 2000, J GEOGRAPHIC INFORMA, V4, P11
[10]   On the Harrison and Rubinfeld data [J].
Gilley, OW ;
Pace, RK .
JOURNAL OF ENVIRONMENTAL ECONOMICS AND MANAGEMENT, 1996, 31 (03) :403-405