𝔖 Bobbio Scriptorium
✦   LIBER   ✦

A nonlinear discriminant analysis

✍ Scribed by N. Victor

Publisher
Elsevier Science
Year
1971
Weight
607 KB
Volume
2
Category
Article
ISSN
0010-468X

No coin nor oath required. For personal study only.

✦ Synopsis

Two linkable computer programs have been developed for a special case of nonlinear discriminant analysis. Here the discriminant formula is nonlinear because joint normal distributions are postulated, but not equal covariance matrices (abbr. CV-matrices). This generalization seems to be important to the computer-aided diagnosis because in biological problems the postulate of equal CV-matrices is never fulfilled practically.

The programs allow us the option to work with or without a priori-probabilities as well as with or without costs. The first program is chiefly useful to estimate all parameters of the discriminant function. The second program makes possible the allocation of new cases as well as the unbiased estimation of error rates.

Discriminant analysis computer-aided diagnosis nonlinear discriminant function unequal covariance matricos

error rates for allocation

πŸ“œ SIMILAR VOLUMES

Comparison of regularized discriminant a
Comparison of regularized discriminant analysis linear discriminant analysis and quadratic discriminant analysis applied to NIR data
✍ W. Wu; Y. Mallet; B. Walczak; W. Penninckx; D.L. Massart; S. Heuerding; F. Erni πŸ“‚ Article πŸ“… 1996 πŸ› Elsevier Science 🌐 English βš– 760 KB

Three classifiers, namely linear discriminant analysis (LDA), quadratic discriminant analysis (QDA) and regularized discriminant analysis (RDA) are considered in this study for classification based on MR data. Because NIR data sets are severely ill-conditioned, the three methods cannot be directly a

A Preliminary Test in Discriminant Analy
A Preliminary Test in Discriminant Analysis
✍ Avner Bar-Hen πŸ“‚ Article πŸ“… 1996 πŸ› Elsevier Science 🌐 English βš– 419 KB

One of the aims of discriminant analysis is the allocation of unknown entities to populations that are known a priori. Consider k populations. Let X denote the vector of observations on an experimental unit, whose origin is uncertain. For the general parametric case, a test is proposed to verify the

A reformative kernel Fisher discriminant
A reformative kernel Fisher discriminant analysis
✍ Yong Xu; Jing-yu Yang; Jian Yang πŸ“‚ Article πŸ“… 2004 πŸ› Elsevier Science 🌐 English βš– 133 KB

A reformative kernel Fisher discriminant method is proposed, which is directly derived from the naive kernel Fisher discriminant analysis with superiority in classiΓΏcation e ciency. In the novel method only a part of training patterns, called "signiΓΏcant nodes", are necessary to be adopted in classi