Muutke küpsiste eelistusi

Statistical Pattern Recognition 3rd edition [Kõva köide]

3.64/5 (15 hinnangut Goodreads-ist)
(QinetiQ Ltd), (QinetiQ Ltd)
  • Formaat: Hardback, 672 pages, kõrgus x laius x paksus: 249x175x41 mm, kaal: 1225 g
  • Ilmumisaeg: 21-Oct-2011
  • Kirjastus: John Wiley & Sons Inc
  • ISBN-10: 0470682272
  • ISBN-13: 9780470682272
Teised raamatud teemal:
Statistical Pattern Recognition 3rd editionSuurem pilt
  • Formaat: Hardback, 672 pages, kõrgus x laius x paksus: 249x175x41 mm, kaal: 1225 g
  • Ilmumisaeg: 21-Oct-2011
  • Kirjastus: John Wiley & Sons Inc
  • ISBN-10: 0470682272
  • ISBN-13: 9780470682272
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9780470682272.html
Märksõnad:
 Statistical Pattern Recognition provides an introduction to statistical pattern theory and techniques, with material drawn from a wide range of fields, including the areas of engineering, statistics, computer science and the social sciences. The book describes techniques for analysing data comprising measurements made on individuals or objects.. The techniques are used to make a prediction such as disease of a patient, the type of object illuminated by a radar, economic forecast. Emphasis is placed on techniques for classification, a term used for predicting the class or group an object belongs to (based on a set of exemplars) and for methods that seek to discover natural groupings in a data set. Each section concludes with a description of the wide range of practical applications that have been addressed and the further developments of theoretical techniques and includes a variety of exercises, from ‘open-book’ questions to more lengthy projects. New material is presented, including the analysis of complex networks and basic techniques for analysing the properties of datasets and also introduces readers to the use of variational methods for Bayesian density estimation and looks at new applications in biometrics and security.  

Arvustused

In the end I must add that this book is so appealing that I often found myself lost in the reading, pausing the overview of the manuscript in order to look more into some presented subject, and not being able to continue until I had finished seeing all about it.  (Zentralblatt MATH, 1 December 2012)

Preface xix
Notation xxiii
1 Introduction to Statistical Pattern Recognition
1(32)
1.1 Statistical Pattern Recognition
1(3)
1.1.1 Introduction
1(1)
1.1.2 The Basic Model
2(2)
1.2 Stages in a Pattern Recognition Problem
4(2)
1.3 Issues
6(1)
1.4 Approaches to Statistical Pattern Recognition
7(1)
1.5 Elementary Decision Theory
8(12)
1.5.1 Bayes' Decision Rule for Minimum Error
8(4)
1.5.2 Bayes' Decision Rule for Minimum Error - Reject Option
12(1)
1.5.3 Bayes' Decision Rule for Minimum Risk
13(2)
1.5.4 Bayes' Decision Rule for Minimum Risk - Reject Option
15(1)
1.5.5 Neyman-Pearson Decision Rule
15(3)
1.5.6 Minimax Criterion
18(1)
1.5.7 Discussion
19(1)
1.6 Discriminant Functions
20(7)
1.6.1 Introduction
20(1)
1.6.2 Linear Discriminant Functions
21(2)
1.6.3 Piecewise Linear Discriminant Functions
23(1)
1.6.4 Generalised Linear Discriminant Function
24(2)
1.6.5 Summary
26(1)
1.7 Multiple Regression
27(2)
1.8 Outline of Book
29(1)
1.9 Notes and References
29(2)
Exercises
31(2)
2 Density Estimation - Parametric
33(37)
2.1 Introduction
33(1)
2.2 Estimating the Parameters of the Distributions
34(1)
2.2.1 Estimative Approach
34(1)
2.2.2 Predictive Approach
35(1)
2.3 The Gaussian Classifier
35(5)
2.3.1 Specification
35(2)
2.3.2 Derivation of the Gaussian Classifier Plug-In Estimates
37(2)
2.3.3 Example Application Study
39(1)
2.4 Dealing with Singularities in the Gaussian Classifier
40(6)
2.4.1 Introduction
40(1)
2.4.2 Naive Bayes
40(1)
2.4.3 Projection onto a Subspace
41(1)
2.4.4 Linear Discriminant Function
41(1)
2.4.5 Regularised Discriminant Analysis
42(2)
2.4.6 Example Application Study
44(1)
2.4.7 Further Developments
45(1)
2.4.8 Summary
46(1)
2.5 Finite Mixture Models
46(17)
2.5.1 Introduction
46(2)
2.5.2 Mixture Models for Discrimination
48(1)
2.5.3 Parameter Estimation for Normal Mixture Models
49(2)
2.5.4 Normal Mixture Model Covariance Matrix Constraints
51(1)
2.5.5 How Many Components?
52(3)
2.5.6 Maximum Likelihood Estimation via EM
55(5)
2.5.7 Example Application Study
60(2)
2.5.8 Further Developments
62(1)
2.5.9 Summary
63(1)
2.6 Application Studies
63(3)
2.7 Summary and Discussion
66(1)
2.8 Recommendations
66(1)
2.9 Notes and References
67(1)
Exercises
67(3)
3 Density Estimation - Bayesian
70(80)
3.1 Introduction
70(3)
3.1.1 Basics
72(1)
3.1.2 Recursive Calculation
72(1)
3.1.3 Proportionality
73(1)
3.2 Analytic Solutions
73(14)
3.2.1 Conjugate Priors
73(2)
3.2.2 Estimating the Mean of a Normal Distribution with Known Variance
75(4)
3.2.3 Estimating the Mean and the Covariance Matrix of a Multivariate Normal Distribution
79(6)
3.2.4 Unknown Prior Class Probabilities
85(2)
3.2.5 Summary
87(1)
3.3 Bayesian Sampling Schemes
87(8)
3.3.1 Introduction
87(1)
3.3.2 Summarisation
87(2)
3.3.3 Sampling Version of the Bayesian Classifier
89(1)
3.3.4 Rejection Sampling
89(1)
3.3.5 Ratio of Uniforms
90(2)
3.3.6 Importance Sampling
92(3)
3.4 Markov Chain Monte Carlo Methods
95(21)
3.4.1 Introduction
95(1)
3.4.2 The Gibbs Sampler
95(8)
3.4.3 Metropolis-Hastings Algorithm
103(4)
3.4.4 Data Augmentation
107(1)
3.4.5 Reversible Jump Markov Chain Monte Carlo
108(1)
3.4.6 Slice Sampling
109(2)
3.4.7 MCMC Example - Estimation of Noisy Sinusoids
111(4)
3.4.8 Summary
115(1)
3.4.9 Notes and References
116(1)
3.5 Bayesian Approaches to Discrimination
116(3)
3.5.1 Labelled Training Data
116(1)
3.5.2 Unlabelled Training Data
117(2)
3.6 Sequential Monte Carlo Samplers
119(7)
3.6.1 Introduction
119(2)
3.6.2 Basic Methodology
121(4)
3.6.3 Summary
125(1)
3.7 Variational Bayes
126(11)
3.7.1 Introduction
126(1)
3.7.2 Description
126(3)
3.7.3 Factorised Variational Approximation
129(2)
3.7.4 Simple Example
131(4)
3.7.5 Use of the Procedure for Model Selection
135(1)
3.7.6 Further Developments and Applications
136(1)
3.7.7 Summary
137(1)
3.8 Approximate Bayesian Computation
137(7)
3.8.1 Introduction
137(1)
3.8.2 ABC Rejection Sampling
138(2)
3.8.3 ABC MCMC Sampling
140(1)
3.8.4 ABC Population Monte Carlo Sampling
141(1)
3.8.5 Model Selection
142(1)
3.8.6 Summary
143(1)
3.9 Example Application Study
144(1)
3.10 Application Studies
145(1)
3.11 Summary and Discussion
146(1)
3.12 Recommendations
147(1)
3.13 Notes and References
147(1)
Exercises
148(2)
4 Density Estimation - Nonparametric
150(71)
4.1 Introduction
150(2)
4.1.1 Basic Properties of Density Estimators
150(2)
4.2 k-Nearest-Neighbour Method
152(28)
4.2.1 k-Nearest-Neighbour Classifier
152(2)
4.2.2 Derivation
154(3)
4.2.3 Choice of Distance Metric
157(2)
4.2.4 Properties of the Nearest-Neighbour Rule
159(1)
4.2.5 Linear Approximating and Eliminating Search Algorithm
159(4)
4.2.6 Branch and Bound Search Algorithms: kd-Trees
163(7)
4.2.7 Branch and Bound Search Algorithms: Ball-Trees
170(4)
4.2.8 Editing Techniques
174(3)
4.2.9 Example Application Study
177(1)
4.2.10 Further Developments
178(1)
4.2.11 Summary
179(1)
4.3 Histogram Method
180(14)
4.3.1 Data Adaptive Histograms
181(1)
4.3.2 Independence Assumption (Naive Bayes)
181(1)
4.3.3 Lancaster Models
182(1)
4.3.4 Maximum Weight Dependence Trees
183(3)
4.3.5 Bayesian Networks
186(4)
4.3.6 Example Application Study - Naive Bayes Text Classification
190(3)
4.3.7 Summary
193(1)
4.4 Kernel Methods
194(10)
4.4.1 Biasedness
197(1)
4.4.2 Multivariate Extension
198(1)
4.4.3 Choice of Smoothing Parameter
199(2)
4.4.4 Choice of Kernel
201(1)
4.4.5 Example Application Study
202(1)
4.4.6 Further Developments
203(1)
4.4.7 Summary
203(1)
4.5 Expansion by Basis Functions
204(3)
4.6 Copulas
207(6)
4.6.1 Introduction
207(1)
4.6.2 Mathematical Basis
207(1)
4.6.3 Copula Functions
208(1)
4.6.4 Estimating Copula Probability Density Functions
209(2)
4.6.5 Simple Example
211(1)
4.6.6 Summary
212(1)
4.7 Application Studies
213(3)
4.7.1 Comparative Studies
216(1)
4.8 Summary and Discussion
216(1)
4.9 Recommendations
217(1)
4.10 Notes and References
217(1)
Exercises
218(3)
5 Linear Discriminant Analysis
221(53)
5.1 Introduction
221(1)
5.2 Two-Class Algorithms
222(14)
5.2.1 General Ideas
222(5)
5.2.3 Fisher's Criterion
227(1)
5.2.4 Least Mean-Squared-Error Procedures
228(7)
5.2.5 Further Developments
235(1)
5.2.6 Summary
235(1)
5.3 Multiclass Algorithms
236(13)
5.3.1 General Ideas
236(1)
5.3.2 Error-Correction Procedure
237(1)
5.3.3 Fisher's Criterion - Linear Discriminant Analysis
238(3)
5.3.4 Least Mean-Squared-Error Procedures
241(5)
5.3.5 Regularisation
246(1)
5.3.6 Example Application Study
246(1)
5.3.7 Further Developments
247(1)
5.3.8 Summary
248(1)
5.4 Support Vector Machines
249(14)
5.4.1 Introduction
249(1)
5.4.2 Linearly Separable Two-Class Data
249(4)
5.4.3 Linearly Nonseparable Two-Class Data
253(3)
5.4.4 Multiclass SVMs
256(1)
5.4.5 SVMs for Regression
257(2)
5.4.6 Implementation
259(3)
5.4.7 Example Application Study
262(1)
5.4.8 Summary
263(1)
5.5 Logistic Discrimination
263(5)
5.5.1 Two-Class Case
263(1)
5.5.2 Maximum Likelihood Estimation
264(2)
5.5.3 Multiclass Logistic Discrimination
266(1)
5.5.4 Example Application Study
267(1)
5.5.5 Further Developments
267(1)
5.5.6 Summary
268(1)
5.6 Application Studies
268(1)
5.7 Summary and Discussion
268(1)
5.8 Recommendations
269(1)
5.9 Notes and References
270(1)
Exercises
270(4)
6 Nonlinear Discriminant Analysis - Kernel and Projection Methods
274(48)
6.1 Introduction
274(2)
6.2 Radial Basis Functions
276(15)
6.2.1 Introduction
276(2)
6.2.2 Specifying the Model
278(1)
6.2.3 Specifying the Functional Form
278(1)
6.2.4 The Positions of the Centres
279(2)
6.2.5 Smoothing Parameters
281(1)
6.2.6 Calculation of the Weights
282(2)
6.2.7 Model Order Selection
284(1)
6.2.8 Simple RBF
285(1)
6.2.9 Motivation
286(2)
6.2.10 RBF Properties
288(1)
6.2.11 Example Application Study
288(1)
6.2.12 Further Developments
289(1)
6.2.13 Summary
290(1)
6.3 Nonlinear Support Vector Machines
291(7)
6.3.1 Introduction
291(1)
6.3.2 Binary Classification
291(1)
6.3.3 Types of Kernel
292(1)
6.3.4 Model Selection
293(1)
6.3.5 Multiclass SVMs
294(1)
6.3.6 Probability Estimates
294(2)
6.3.7 Nonlinear Regression
296(1)
6.3.8 Example Application Study
296(1)
6.3.9 Further Developments
297(1)
6.3.10 Summary
298(1)
6.4 The Multilayer Perceptron
298(16)
6.4.1 Introduction
298(1)
6.4.2 Specifying the MLP Structure
299(1)
6.4.3 Determining the MLP Weights
300(7)
6.4.4 Modelling Capacity of the MLP
307(1)
6.4.5 Logistic Classification
307(3)
6.4.6 Example Application Study
310(1)
6.4.7 Bayesian MLP Networks
311(2)
6.4.8 Projection Pursuit
313(1)
6.4.9 Summary
313(1)
6.5 Application Studies
314(2)
6.6 Summary and Discussion
316(1)
6.7 Recommendations
317(1)
6.8 Notes and References
318(1)
Exercises
318(4)
7 Rule and Decision Tree Induction
322(39)
7.1 Introduction
322(1)
7.2 Decision Trees
323(19)
7.2.1 Introduction
323(3)
7.2.2 Decision Tree Construction
326(1)
7.2.3 Selection of the Splitting Rule
327(3)
7.2.4 Terminating the Splitting Procedure
330(2)
7.2.5 Assigning Class Labels to Terminal Nodes
332(1)
7.2.6 Decision Tree Pruning - Worked Example
332(5)
7.2.7 Decision Tree Construction Methods
337(2)
7.2.8 Other Issues
339(1)
7.2.9 Example Application Study
340(1)
7.2.10 Further Developments
341(1)
7.2.11 Summary
342(1)
7.3 Rule Induction
342(9)
7.3.1 Introduction
342(3)
7.3.2 Generating Rules from a Decision Tree
345(1)
7.3.3 Rule Induction Using a Sequential Covering Algorithm
345(5)
7.3.4 Example Application Study
350(1)
7.3.5 Further Developments
351(1)
7.3.6 Summary
351(1)
7.4 Multivariate Adaptive Regression Splines
351(5)
7.4.1 Introduction
351(1)
7.4.2 Recursive Partitioning Model
351(4)
7.4.3 Example Application Study
355(1)
7.4.4 Further Developments
355(1)
7.4.5 Summary
356(1)
7.5 Application Studie
356(2)
7.6 Summary and Discussion
358(1)
7.7 Recommendations
358(1)
7.8 Notes and References
359(1)
Exercises
359(2)
8 Ensemble Methods
361(43)
8.1 Introduction
361(1)
8.2 Characterising a Classifier Combination Scheme
362(8)
8.2.1 Feature Space
363(3)
8.2.2 Level
366(2)
8.2.3 Degree of Training
368(1)
8.2.4 Form of Component Classifiers
368(1)
8.2.5 Structure
369(1)
8.2.6 Optimisation
369(1)
8.3 Data Fusion
370(6)
8.3.1 Architectures
370(1)
8.3.2 Bayesian Approaches
371(2)
8.3.3 Neyman-Pearson Formulation
373(1)
8.3.4 Trainable Rules
374(1)
8.3.5 Fixed Rules
375(1)
8.4 Classifier Combination Methods
376(23)
8.4.1 Product Rule
376(1)
8.4.2 Sum Rule
377(1)
8.4.3 Min, Max and Median Combiners
378(1)
8.4.4 Majority Vote
379(1)
8.4.5 Borda Count
379(1)
8.4.6 Combiners Trained on Class Predictions
380(2)
8.4.7 Stacked Generalisation
382(1)
8.4.8 Mixture of Experts
382(3)
8.4.9 Bagging
385(2)
8.4.10 Boosting
387(2)
8.4.11 Random Forests
389(1)
8.4.12 Model Averaging
390(6)
8.4.13 Summary of Methods
396(2)
8.4.14 Example Application Study
398(1)
8.4.15 Further Developments
399(1)
8.5 Application Studies
399(1)
8.6 Summary and Discussion
400(1)
8.7 Recommendations
401(1)
8.8 Notes and References
401(1)
Exercises
402(2)
9 Performance Assessment
404(29)
9.1 Introduction
404(1)
9.2 Performance Assessment
405(19)
9.2.1 Performance Measures
405(1)
9.2.2 Discriminability
406(7)
9.2.3 Reliability
413(2)
9.2.4 ROC Curves for Performance Assessment
415(4)
9.2.5 Population and Sensor Drift
419(2)
9.2.6 Example Application Study
421(1)
9.2.7 Further Developments
422(1)
9.2.8 Summary
423(1)
9.3 Comparing Classifier Performance
424(5)
9.3.1 Which Technique is Best?
424(1)
9.3.2 Statistical Tests
425(1)
9.3.3 Comparing Rules When Misclassification Costs are Uncertain
426(2)
9.3.4 Example Application Study
428(1)
9.3.5 Further Developments
429(1)
9.3.6 Summary
429(1)
9.4 Application Studies
429(1)
9.5 Summary and Discussion
430(1)
9.6 Recommendations
430(1)
9.7 Notes and References
430(1)
Exercises
431(2)
10 Feature Selection and Extraction
433(68)
10.1 Introduction
433(2)
10.2 Feature Selection
435(28)
10.2.1 Introduction
435(4)
10.2.2 Characterisation of Feature Selection Approaches
439(1)
10.2.3 Evaluation Measures
440(9)
10.2.4 Search Algorithms for Feature Subset Selection
449(1)
10.2.5 Complete Search - Branch and Bound
450(4)
10.2.6 Sequential Search
454(4)
10.2.7 Random Search
458(1)
10.2.8 Markov Blanket
459(1)
10.2.9 Stability of Feature Selection
460(2)
10.2.10 Example Application Study
462(1)
10.2.11 Further Developments
462(1)
10.2.12 Summary
463(1)
10.3 Linear Feature Extraction
463(21)
10.3.1 Principal Components Analysis
464(11)
10.3.2 Karhunen-Loeve Transformation
475(6)
10.3.3 Example Application Study
481(1)
10.3.4 Further Developments
482(1)
10.3.5 Summary
483(1)
10.4 Multidimensional Scaling
484(9)
10.4.1 Classical Scaling
484(2)
10.4.2 Metric MDS
486(1)
10.4.3 Ordinal Scaling
487(3)
10.4.4 Algorithms
490(1)
10.4.5 MDS for Feature Extraction
491(1)
10.4.6 Example Application Study
492(1)
10.4.7 Further Developments
493(1)
10.4.8 Summary
493(1)
10.5 Application Studies
493(2)
10.6 Summary and Discussion
495(1)
10.7 Recommendations
495(1)
10.8 Notes and References
496(1)
Exercises
497(4)
11 Clustering
501(54)
11.1 Introduction
501(1)
11.2 Hierarchical Methods
502(8)
11.2.1 Single-Link Method
503(3)
11.2.2 Complete-Link Method
506(1)
11.2.3 Sum-of-Squares Method
507(1)
11.2.4 General Agglomerative Algorithm
508(1)
11.2.5 Properties of a Hierarchical Classification
508(1)
11.2.6 Example Application Study
509(1)
11.2.7 Summary
509(1)
11.3 Quick Partitions
510(1)
11.4 Mixture Models
511(2)
11.4.1 Model Description
511(1)
11.4.2 Example Application Study
512(1)
11.5 Sum-of-Squares Methods
513(18)
11.5.1 Clustering Criteria
514(1)
11.5.2 Clustering Algorithms
515(5)
11.5.3 Vector Quantisation
520(10)
11.5.4 Example Application Study
530(1)
11.5.5 Further Developments
530(1)
11.5.6 Summary
531(1)
11.6 Spectral Clustering
531(7)
11.6.1 Elementary Graph Theory
531(3)
11.6.2 Similarity Matrices
534(1)
11.6.3 Application to Clustering
534(1)
11.6.4 Spectral Clustering Algorithm
535(1)
11.6.5 Forms of Graph Laplacian
535(1)
11.6.6 Example Application Study
536(2)
11.6.7 Further Developments
538(1)
11.6.8 Summary
538(1)
11.7 Cluster Validity
538(8)
11.7.1 Introduction
538(1)
11.7.2 Statistical Tests
539(1)
11.7.3 Absence of Class Structure
540(1)
11.7.4 Validity of Individual Clusters
541(1)
11.7.5 Hierarchical Clustering
542(1)
11.7.6 Validation of Individual Clusterings
542(1)
11.7.7 Partitions
543(1)
11.7.8 Relative Criteria
543(2)
11.7.9 Choosing the Number of Clusters
545(1)
11.8 Application Studies
546(3)
11.9 Summary and Discussion
549(2)
11.10 Recommendations
551(1)
11.11 Notes and References
552(1)
Exercises
553(2)
12 Complex Networks
555(26)
12.1 Introduction
555(6)
12.1.1 Characteristics
557(1)
12.1.2 Properties
557(2)
12.1.3 Questions to Address
559(1)
12.1.4 Descriptive Features
560(1)
12.1.5 Outline
560(1)
12.2 Mathematics of Networks
561(4)
12.2.1 Graph Matrices
561(1)
12.2.2 Connectivity
562(1)
12.2.3 Distance Measures
562(1)
12.2.4 Weighted Networks
563(1)
12.2.5 Centrality Measures
563(1)
12.2.6 Random Graphs
564(1)
12.3 Community Detection
565(10)
12.3.1 Clustering Methods
565(3)
12.3.2 Girvan-Newman Algorithm
568(2)
12.3.3 Modularity Approaches
570(1)
12.3.4 Local Modularity
571(2)
12.3.5 Clique Percolation
573(1)
12.3.6 Example Application Study
574(1)
12.3.7 Further Developments
575(1)
12.3.8 Summary
575(1)
12.4 Link Prediction
575(4)
12.4.1 Approaches to Link Prediction
576(2)
12.4.2 Example Application Study
578(1)
12.4.3 Further Developments
578(1)
12.5 Application Studies
579(1)
12.6 Summary and Discussion
579(1)
12.7 Recommendations
580(1)
12.8 Notes and References
580(1)
Exercises
580(1)
13 Additional Topics
581(10)
13.1 Model Selection
581(4)
13.1.1 Separate Training and Test Sets
582(1)
13.1.2 Cross-Validation
582(1)
13.1.3 The Bayesian Viewpoint
583(1)
13.1.4 Akaike's Information Criterion
583(1)
13.1.5 Minimum Description Length
584(1)
13.2 Missing Data
585(1)
13.3 Outlier Detection and Robust Procedures
586(1)
13.4 Mixed Continuous and Discrete Variables
587(1)
13.5 Structural Risk Minimisation and the Vapnik-Chervonenkis Dimension
588(3)
13.5.1 Bounds on the Expected Risk
588(1)
13.5.2 The VC Dimension
589(2)
References 591(46)
Index 637
Dr Andrew Robert Webb, Senior Researcher, QinetiQ Ltd, Malvern, UK.

Dr Keith Derek Copsey, Senior Researcher, QinetiQ Ltd, Malvern, UK.