Discrimination and ClassificationPresents different approaches to discrimination and classification problems from a statistical perspective. Provides computer projects concentrating on the most widely used and important algorithms, numerical examples, and theoretical questions reinforce to further develop the ideas introduced in the text. |
Contents
DistributionFree Methods | 16 |
Parameterized Distributions | 45 |
8 | 71 |
Copyright | |
9 other sections not shown
Other editions - View all
Common terms and phrases
algorithm applied approach assumptions asymptotic Bayes binary variables branch and bound categorical variables cells Chapter class-conditional pdfs classification rule classifier cluster analysis coefficients complete space components continuous variables converge data set decision rule defined design set points dimensionality discriminant analysis discriminant function discussed distance evaluated example Figure Fisher's linear discriminant Fukunaga IEEE Transactions incomplete vectors Information Theory iterative kernel method Lachenbruch large number leaving-one-out method linear discriminant function linear programming Math matrix maximum likelihood McLachlan mean measure minimize misclassification rate multinomial multivariate normal n₁ n₂ nearest neighbour node non-parametric normal distribution objects observations optimal parameters partition pattern recognition pdf estimation perceptron problem procedure regression sample points Section separability solution statistical stochastic approximation subset subspace sum of squares techniques Technometrics Transactions on Information transformation v₁ values variable selection variable set variance-covariance matrices variance-covariance matrix w₁ w₂ Wk+1 x₁