Pattern classification

>> Google Books

所蔵情報QRコード

Pattern classification / Richard O. Duda, Peter E. Hart, David G. Stork

版:

2nd ed

資料種別:

図書

出版情報:

New York ; Chichester : Wiley, c2001

形態:

xx, 654 p. ; 27 cm

著者名:

ISBN:

9780471056690 [0471056693]

書誌ID:

BA4963402X

子書誌情報

フルテキスト

所蔵情報

他の版・巻

書誌詳細

その他の標題:

Pattern classification and scene analysis

注記:

"A Wiley-Interscience publication"
Rev. ed. of: Pattern classification and scene analysis / R.O. Duda and P.E. Hart (Wiley-Interscience, c1973)
Includes bibliographical references and index

主題:

Pattern recognition systems; Statistical decision

言語:

英語

目次情報:

Preface

Introduction / 1：

Machine Perception / 1.1：

An Example / 1.2：

Related Fields / 1.2.1：

Pattern Recognition Systems / 1.3：

Sensing / 1.3.1：

Segmentation and Grouping / 1.3.2：

Feature Extraction / 1.3.3：

Classification / 1.3.4：

Post Processing / 1.3.5：

The Design Cycle / 1.4：

Data Collection / 1.4.1：

Feature Choice / 1.4.2：

Model Choice / 1.4.3：

Training / 1.4.4：

Evaluation / 1.4.5：

Computational Complexity / 1.4.6：

Learning and Adaptation / 1.5：

Supervised Learning / 1.5.1：

Unsupervised Learning / 1.5.2：

Reinforcement Learning / 1.5.3：

Conclusion / 1.6：

Summary by Chapters

Bibliographical and Historical Remarks

Bibliography

Bayesian Decision Theory / 2：

Bayesian Decision Theory--Continuous Features / 2.1：

Two-Category Classification / 2.2.1：

Minimum-Error-Rate Classification / 2.3：

Minimax Criterion / 2.3.1：

Neyman-Pearson Criterion / 2.3.2：

Classifiers, Discriminant Functions, and Decision Surfaces / 2.4：

The Multicategory Case / 2.4.1：

The Two-Category Case / 2.4.2：

The Normal Density / 2.5：

Univariate Density / 2.5.1：

Multivariate Density / 2.5.2：

Discriminant Functions for the Normal Density / 2.6：

Case 1: [Sigma subscript i] = [sigma superscript 2]I / 2.6.1：

Case 2: [Sigma ubscript i] = [Sigma] / 2.6.2：

Case 3: [Sigma subscript i] = arbitrary / 2.6.3：

Decision Regions for Two-Dimensional Gaussian Data / Example 1：

Error Probabilities and Integrals / 2.7：

Error Bounds for Normal Densities / 2.8：

Chernoff Bound / 2.8.1：

Bhattacharyya Bound / 2.8.2：

Error Bounds for Gaussian Distributions / Example 2：

Signal Detection Theory and Operating Characteristics / 2.8.3：

Bayes Decision Theory--Discrete Features / 2.9：

Independent Binary Features / 2.9.1：

Bayesian Decisions for Three-Dimensional Binary Data / Example 3：

Missing and Noisy Features / 2.10：

Missing Features / 2.10.1：

Noisy Features / 2.10.2：

Bayesian Belief Networks / 2.11：

Belief Network for Fish / Example 4：

Compound Bayesian Decision Theory and Context / 2.12：

Summary

Problems

Computer exercises

Maximum-Likelihood and Bayesian Parameter Estimation / 3：

Maximum-Likelihood Estimation / 3.1：

The General Principle / 3.2.1：

The Gaussian Case: Unknown [mu] / 3.2.2：

The Gaussian Case: Unknown [mu] and [Sigma] / 3.2.3：

Bias / 3.2.4：

Bayesian Estimation / 3.3：

The Class-Conditional Densities / 3.3.1：

The Parameter Distribution / 3.3.2：

Bayesian Parameter Estimation: Gaussian Case / 3.4：

The Univariate Case: p([mu]|D) / 3.4.1：

The Univariate Case: p(x|D) / 3.4.2：

The Multivariate Case / 3.4.3：

Bayesian Parameter Estimation: General Theory / 3.5：

Recursive Bayes Learning

When Do Maximum-Likelihood and Bayes Methods Differ? / 3.5.1：

Noninformative Priors and Invariance / 3.5.2：

Gibbs Algorithm / 3.5.3：

Sufficient Statistics / 3.6：

Sufficient Statistics and the Exponential Family / 3.6.1：

Problems of Dimensionality / 3.7：

Accuracy, Dimension, and Training Sample Size / 3.7.1：

Overfitting / 3.7.2：

Component Analysis and Discriminants / 3.8：

Principal Component Analysis (PCA) / 3.8.1：

Fisher Linear Discriminant / 3.8.2：

Multiple Discriminant Analysis / 3.8.3：

Expectation-Maximization (EM) / 3.9：

Expectation-Maximization for a 2D Normal Model

Hidden Markov Models / 3.10：

First-Order Markov Models / 3.10.1：

First-Order Hidden Markov Models / 3.10.2：

Hidden Markov Model Computation / 3.10.3：

Hidden Markov Model / 3.10.4：

Decoding / 3.10.5：

HMM Decoding

Learning / 3.10.6：

Nonparametric Techniques / 4：

Density Estimation / 4.1：

Parzen Windows / 4.3：

Convergence of the Mean / 4.3.1：

Convergence of the Variance / 4.3.2：

Illustrations / 4.3.3：

Classification Example / 4.3.4：

Probabilistic Neural Networks (PNNs) / 4.3.5：

Choosing the Window Function / 4.3.6：

k[subscript n]-Nearest-Neighbor Estimation / 4.4：

k[subscript n]-Nearest-Neighbor and Parzen-Window Estimation / 4.4.1：

Estimation of A Posteriori Probabilities / 4.4.2：

The Nearest-Neighbor Rule / 4.5：

Convergence of the Nearest Neighbor / 4.5.1：

Error Rate for the Nearest-Neighbor Rule / 4.5.2：

Error Bounds / 4.5.3：

The k-Nearest-Neighbor Rule / 4.5.4：

Computational Complexity of the k-Nearest-Neighbor Rule / 4.5.5：

Metrics and Nearest-Neighbor Classification / 4.6：

Properties of Metrics / 4.6.1：

Tangent Distance / 4.6.2：

Fuzzy Classification / 4.7：

Reduced Coulomb Energy Networks / 4.8：

Approximations by Series Expansions / 4.9：

Linear Discriminant Functions / 5：

Linear Discriminant Functions and Decision Surfaces / 5.1：

Generalized Linear Discriminant Functions / 5.2.1：

The Two-Category Linearly Separable Case / 5.4：

Geometry and Terminology / 5.4.1：

Gradient Descent Procedures / 5.4.2：

Minimizing the Perceptron Criterion Function / 5.5：

The Perceptron Criterion Function / 5.5.1：

Convergence Proof for Single-Sample Correction / 5.5.2：

Some Direct Generalizations / 5.5.3：

Relaxation Procedures / 5.6：

The Descent Algorithm / 5.6.1：

Convergence Proof / 5.6.2：

Nonseparable Behavior / 5.7：

Minimum Squared-Error Procedures / 5.8：

Minimum Squared-Error and the Pseudoinverse / 5.8.1：

Constructing a Linear Classifier by Matrix Pseudoinverse

Relation to Fisher's Linear Discriminant / 5.8.2：

Asymptotic Approximation to an Optimal Discriminant / 5.8.3：

The Widrow-Hoff or LMS Procedure / 5.8.4：

Stochastic Approximation Methods / 5.8.5：

The Ho-Kashyap Procedures / 5.9：

The Descent Procedure / 5.9.1：

Some Related Procedures / 5.9.2：

Linear Programming Algorithms / 5.10：

Linear Programming / 5.10.1：

The Linearly Separable Case / 5.10.2：

Support Vector Machines / 5.10.3：

SVM Training / 5.11.1：

SVM for the XOR Problem

Multicategory Generalizations / 5.12：

Kesler's Construction / 5.12.1：

Convergence of the Fixed-Increment Rule / 5.12.2：

Generalizations for MSE Procedures / 5.12.3：

Multilayer Neural Networks / 6：

Feedforward Operation and Classification / 6.1：

General Feedforward Operation / 6.2.1：

Expressive Power of Multilayer Networks / 6.2.2：

Backpropagation Algorithm / 6.3：

Network Learning / 6.3.1：

Training Protocols / 6.3.2：

Learning Curves / 6.3.3：

Error Surfaces / 6.4：

Some Small Networks / 6.4.1：

The Exclusive-OR (XOR) / 6.4.2：

Larger Networks / 6.4.3：

How Important Are Multiple Minima? / 6.4.4：

Backpropagation as Feature Mapping / 6.5：

Representations at the Hidden Layer--Weights / 6.5.1：

Backpropagation, Bayes Theory and Probability / 6.6：

Bayes Discriminants and Neural Networks / 6.6.1：

Outputs as Probabilities / 6.6.2：

Related Statistical Techniques / 6.7：

Practical Techniques for Improving Backpropagation / 6.8：

Activation Function / 6.8.1：

Parameters for the Sigmoid / 6.8.2：

Scaling Input / 6.8.3：

Target Values / 6.8.4：

Training with Noise / 6.8.5：

Manufacturing Data / 6.8.6：

Number of Hidden Units / 6.8.7：

Initializing Weights / 6.8.8：

Learning Rates / 6.8.9：

Momentum / 6.8.10：

Weight Decay / 6.8.11：

Hints / 6.8.12：

On-Line, Stochastic or Batch Training? / 6.8.13：

Stopped Training / 6.8.14：

Number of Hidden Layers / 6.8.15：

Criterion Function / 6.8.16：

Second-Order Methods / 6.9：

Hessian Matrix / 6.9.1：

Newton's Method / 6.9.2：

Quickprop / 6.9.3：

Conjugate Gradient Descent / 6.9.4：

Additional Networks and Training Methods / 6.10：

Radial Basis Function Networks (RBFs) / 6.10.1：

Special Bases / 6.10.2：

Matched Filters / 6.10.3：

Convolutional Networks / 6.10.4：

Recurrent Networks / 6.10.5：

Cascade-Correlation / 6.10.6：

Regularization, Complexity Adjustment and Pruning / 6.11：

Stochastic Methods / 7：

Stochastic Search / 7.1：

Simulated Annealing / 7.2.1：

The Boltzmann Factor / 7.2.2：

Deterministic Simulated Annealing / 7.2.3：

Boltzmann Learning / 7.3：

Stochastic Boltzmann Learning of Visible States / 7.3.1：

Missing Features and Category Constraints / 7.3.2：

Deterministic Boltzmann Learning / 7.3.3：

Initialization and Setting Parameters / 7.3.4：

Boltzmann Networks and Graphical Models / 7.4：

Other Graphical Models / 7.4.1：

Evolutionary Methods / 7.5：

Genetic Algorithms / 7.5.1：

Further Heuristics / 7.5.2：

Why Do They Work? / 7.5.3：

Genetic Programming / 7.6：

Nonmetric Methods / 8：

Decision Trees / 8.1：

Cart / 8.3：

Number of Splits / 8.3.1：

Query Selection and Node Impurity / 8.3.2：

When to Stop Splitting / 8.3.3：

Pruning / 8.3.4：

Assignment of Leaf Node Labels / 8.3.5：

A Simple Tree

Multivariate Decision Trees / 8.3.6：

Priors and Costs / 8.3.9：

Missing Attributes / 8.3.10：

Surrogate Splits and Missing Attributes

Other Tree Methods / 8.4：

ID3 / 8.4.1：

C4.5 / 8.4.2：

Which Tree Classifier Is Best? / 8.4.3：

Recognition with Strings / 8.5：

String Matching / 8.5.1：

Edit Distance / 8.5.2：

String Matching with Errors / 8.5.3：

String Matching with the "Don't-Care" Symbol / 8.5.5：

Grammatical Methods / 8.6：

Grammars / 8.6.1：

Types of String Grammars / 8.6.2：

A Grammar for Pronouncing Numbers

Recognition Using Grammars / 8.6.3：

Grammatical Inference / 8.7：

Rule-Based Methods / 8.8：

Learning Rules / 8.8.1：

Algorithm-Independent Machine Learning / 9：

Lack of Inherent Superiority of Any Classifier / 9.1：

No Free Lunch Theorem / 9.2.1：

No Free Lunch for Binary Data

Ugly Duckling Theorem / 9.2.2：

Minimum Description Length (MDL) / 9.2.3：

Minimum Description Length Principle / 9.2.4：

Overfitting Avoidance and Occam's Razor / 9.2.5：

Bias and Variance / 9.3：

Bias and Variance for Regression / 9.3.1：

Bias and Variance for Classification / 9.3.2：

Resampling for Estimating Statistics / 9.4：

Jackknife / 9.4.1：

Jackknife Estimate of Bias and Variance of the Mode

Bootstrap / 9.4.2：

Resampling for Classifier Design / 9.5：

Bagging / 9.5.1：

Boosting / 9.5.2：

Learning with Queries / 9.5.3：

Arcing, Learning with Queries, Bias and Variance / 9.5.4：

Estimating and Comparing Classifiers / 9.6：

Parametric Models / 9.6.1：

Cross-Validation / 9.6.2：

Jackknife and Bootstrap Estimation of Classification Accuracy / 9.6.3：

Maximum-Likelihood Model Comparison / 9.6.4：

Bayesian Model Comparison / 9.6.5：

The Problem-Average Error Rate / 9.6.6：

Predicting Final Performance from Learning Curves / 9.6.7：

The Capacity of a Separating Plane / 9.6.8：

Combining Classifiers / 9.7：

Component Classifiers with Discriminant Functions / 9.7.1：

Component Classifiers without Discriminant Functions / 9.7.2：

Unsupervised Learning and Clustering / 10：

Mixture Densities and Identifiability / 10.1：

Maximum-Likelihood Estimates / 10.3：

Application to Normal Mixtures / 10.4：

Case 1: Unknown Mean Vectors / 10.4.1：

Case 2: All Parameters Unknown / 10.4.2：

k-Means Clustering / 10.4.3：

Fuzzy k-Means Clustering / 10.4.4：

Unsupervised Bayesian Learning / 10.5：

The Bayes Classifier / 10.5.1：

Learning the Parameter Vector / 10.5.2：

Unsupervised Learning of Gaussian Data

Decision-Directed Approximation / 10.5.3：

Data Description and Clustering / 10.6：

Similarity Measures / 10.6.1：

Criterion Functions for Clustering / 10.7：

The Sum-of-Squared-Error Criterion / 10.7.1：

Related Minimum Variance Criteria / 10.7.2：

Scatter Criteria / 10.7.3：

Clustering Criteria

Iterative Optimization / 10.8：

Hierarchical Clustering / 10.9：

Definitions / 10.9.1：

Agglomerative Hierarchical Clustering / 10.9.2：

Stepwise-Optimal Hierarchical Clustering / 10.9.3：

Hierarchical Clustering and Induced Metrics / 10.9.4：

The Problem of Validity / 10.10：

On-line clustering / 10.11：

Unknown Number of Clusters / 10.11.1：

Adaptive Resonance / 10.11.2：

Learning with a Critic / 10.11.3：

Graph-Theoretic Methods / 10.12：

Component Analysis / 10.13：

Nonlinear Component Analysis (NLCA) / 10.13.1：

Independent Component Analysis (ICA) / 10.13.3：

Low-Dimensional Representations and Multidimensional Scaling (MDS) / 10.14：

Self-Organizing Feature Maps / 10.14.1：

Clustering and Dimensionality Reduction / 10.14.2：

Mathematical Foundations / A：

Notation / A.1：

Linear Algebra / A.2：

Notation and Preliminaries / A.2.1：

Inner Product / A.2.2：

Outer Product / A.2.3：

Derivatives of Matrices / A.2.4：

Determinant and Trace / A.2.5：

Matrix Inversion / A.2.6：

Eigenvectors and Eigenvalues / A.2.7：

Lagrange Optimization / A.3：

Probability Theory / A.4：

Discrete Random Variables / A.4.1：

Expected Values / A.4.2：

Pairs of Discrete Random Variables / A.4.3：

Statistical Independence / A.4.4：

Expected Values of Functions of Two Variables / A.4.5：

Conditional Probability / A.4.6：

The Law of Total Probability and Bayes' Rule / A.4.7：

Vector Random Variables / A.4.8：

Expectations, Mean Vectors and Covariance Matrices / A.4.9：

Continuous Random Variables / A.4.10：

Distributions of Sums of Independent Random Variables / A.4.11：

Normal Distributions / A.4.12：

Gaussian Derivatives and Integrals / A.5：

Multivariate Normal Densities / A.5.1：

Bivariate Normal Densities / A.5.2：

Hypothesis Testing / A.6：

Chi-Squared Test / A.6.1：

Information Theory / A.7：

Entropy and Information / A.7.1：

Relative Entropy / A.7.2：

Mutual Information / A.7.3：

Index / A.8：

Preface

Introduction / 1：

Machine Perception / 1.1：

An Example / 1.2：

Related Fields / 1.2.1：

Pattern Recognition Systems / 1.3：

続きを見る

ブックレビュー

類似資料:

1 図書 Pattern classification and scene analysis Duda, Richard O., Hart, Peter E. Wiley	7 図書 Pattern recognition using neural networks : theory and algorithms for engineers and scientists (: cl) Looney, Carl G. Oxford University Press
2 図書 Pattern classification : a unified view of statistical and neural approaches Schürmann, Jürgen Wiley	8 図書 Risk in business decision Moore, P. G. (Peter Gerald), 1928- Longman
3 図書パターン識別 Duda, Richard O., Hart, Peter E., Stork, David G., 尾上, 守夫新技術コミュニケーションズ	9 図書 Pattern synthesis Grenander, Ulf, 1923- Springer-Verlag
4 図書 Practical approach to pattern classification Batchelor, Bruce G. Plenum Press	10 図書 Pattern analysis (: gw) Grenander, Ulf, 1923- Springer-Verlag
5 図書 Recognition of patterns : using the frequencies of occurrence of binary words (: Wien ; : New York) Becker, Peter W. Springer-Verlag	11 図書 Machine recognition of patterns (: cloth : IEEE - : paper : Wiley) Agrawala, Ashok K. IEEE Press : sole worldwide distributor (exclusive of IEEE), Wiley
6 図書 The physics of sound Berg, Richard E., Stork, David G. Prentice-Hall	12 図書 Neural networks and pattern recognition Omidvar, Omid, Dayhoff, Judith E. Academic Press