Entropy-based subspace clustering for mining numerical data
KDD '99 Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining
An introduction to support Vector Machines: and other kernel-based learning methods
An introduction to support Vector Machines: and other kernel-based learning methods
Spartan Gibbs Random Field Models for Geostatistical Applications
SIAM Journal on Scientific Computing
Automatic Subspace Clustering of High Dimensional Data
Data Mining and Knowledge Discovery
IEEE Transactions on Signal Processing - Part I
Analytic Properties and Covariance Functions for a New Class of Generalized Gibbs Random Fields
IEEE Transactions on Information Theory
Prototyping the visualization of geographic and sensor data for agriculture
Computers and Electronics in Agriculture
Hi-index | 0.00 |
This paper addresses the estimation of geometric anisotropy parameters from scattered spatial data that are obtained from environmental surveillance networks. Estimates of geometric anisotropy improve the accuracy of spatial interpolation procedures that aim to generate smooth maps for visualization of the data and for decision making purposes. The anisotropy parameters involve the orientation angle of the principal anisotropy axes and the anisotropy ratio (i.e., the ratio of the principal correlation lengths). The approach that we employ is based on the covariance Hessian identity (CHI) method, which links the mean gradient tensor with the Hessian matrix of the covariance function. We extend CHI to clustered CHI for application in data sets that include patches of extreme values and clusters of varying sampling density. We investigate the impact of CHI anisotropy estimation on the performance of spatial interpolation by ordinary kriging using a data set that involves both real background radioactivity measurements and a simulated release of a radioactive plume.