Efficient implementation of some contextual classification methods

MaRDI QIDQ1965995zbMATH OpenFDO

Authors A.-M. Flygare, Anders Barrlund

Publication date 2 March 2000

Published in Computational Statistics (Search for Journal in Brave)

zbMATH Keywords

autocorrelationCholesky factorizationcontextual classification modelsspatial dependence

Mathematics Subject Classification ID

Factorization of matrices (15A23) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Inference from spatial processes (62M30)

Review text

The authors describe how a special form of the Cholesky factorization of the Kronecker product \(A\otimes \Sigma\) of two symmetric and positive definite matrices \(A\) and \(\Sigma\) can be used to efficiently compute \(x^T \left(A\otimes\Sigma\right)^{-1}x.\) The method is then used to develop an efficient algorithm to implement contextual classification models. An algorithm for their implementation is presented. The efficiency of this algorithm is analysed.

Recommendations

• Algorithm 993: Efficient computation with Kronecker products
• Simple algorithms about Kronecker products in the linear model
• Efficient algorithm for a class of least squares estimation problems
• scientific article; zbMATH DE number 741107
• Computationally efficient cholesky factorization of a covariance matrix with block toeplitz structure

Cited in

(2)

• Efficient exemplars for classifier design
• Efficient Contextual Unfolding

This page was built for publication: Efficient implementation of some contextual classification methods

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1965995)