On optimal pairwise linear classifiers for normal distributions: The d-dimensional case

Pattern Recognition

Volumn 36, Issue 1, 2003, Pages 13-23

  Rueda, Luis G ^a   Oommen, B John ^a  

^a CARLETON UNIVERSITY   (Canada)

Author keywords

Bayesian classification; Linear classifiers; Medical diagnosis; Minsky's paradox; Multi dimensional normal distributions; Optimal classifiers

Indexed keywords

CLASSIFICATION (OF INFORMATION); MATRIX ALGEBRA; MAXIMUM LIKELIHOOD ESTIMATION; NEURAL NETWORKS; OPTIMAL SYSTEMS; VECTORS;

LINEAR CLASSIFIERS;

PATTERN RECOGNITION;

IMAGE CLASSIFICATION; PATTERN RECOGNITION;

EID: 0037209417     PISSN: 00313203     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0031-3203(02)00053-5     Document Type: Article

References (20)

1. On optimal pairwise linear classifiers for normal distributions: The two-dimensional case
2. On an extended fisher criterion for feature selection
3. An introduction to computing with neural nets
4. Nearest neighbor pattern classification perceptrons
5. Evolution and generalization of a single neurone: I. Single-layer perception as seven statistical classifiers
6. Evolution and generalization of a single neurone: II. Complexity of statistical classifiers and sample size considerations
7. A deterministic annealing approach for parsimonious design of piecewise regression models
8. On dimensionality, sample size, and classification error of nonparametric linear classification
9. Linear discriminant analysis for two classes via removal of classification structure
10. Adaptive linear dinensionality reduction for classification
11. A linear constrained distance-based discriminant analysis for hyperspectral image classification
12. The foundational theory of optimal Bayesian pairwise linear classifiers
13. Discriminant analysis with singular covariance matrices: Methods and applications to spectroscopic data
14. Breast cancer diagnosis and prognosis via linear programming