模式识别与神经网络
图书信息
书 名: 模式识别与神经网络
作者：（英）里普利
出版社：人民邮电出版社
出版时间： 2009-8-1
ISBN: 9787115210647
开本： 16开
定价: 69.00元
内容简介本书是模式识别和神经网络方面的名著，讲述了模式识别所涉及的统计方法、神经网络和机器学习等分支。书的内容从介绍和例子开始，主要涵盖统计决策理论、线性判别分析、弹性判别分析、前馈神经网络、非参数方法、树结构分类、信念网、无监管方法、探寻优良的模式特性等方面的内容。
本书可作为统计与理工科研究生课程的教材，对模式识别和神经网络领域的研究人员也是极有价值的参考书。
作者简介里普利（Ripley）著名的统计学家，牛津大学应用统计教授。他在空间统计学、模式识别领域作出了重要贡献，对S的开发以及S-PLUSUS和R的推广应用有着重要影响。20世纪90年代他出版了人工神经网络方面的著作，影响很大，引导统计学者开始关注机器学习和数据挖掘。除本书外，他还著有Modern Applied Statistics with S和S Programming。
图书目录1Introduction and Examples
1.1How do neural methods differ?
1.2The patterm recognition task
1.3Overview of the remaining chapters
1.4Examples
1.5Literature
2Statistical Decision Theory
2.1Bayes rules for known distributions
2.2Parametric models
2.3Logistic discrimination
2.4Predictive classification
2.5Alternative estimation procedures
2.6How complex a model do we need?
2.7Performance assessment
2.8Computational learning approaches
3Linear Discriminant Analysis
3.1Classical linear discriminatio
3.2Linear discriminants via regression
3.3Robustness
3.4Shrinkage methods
3.5Logistic discrimination
3.6Linear separatio andperceptrons
4Flexible Diseriminants
4.1Fitting smooth parametric functions
4.2Radial basis functions
4.3Regularization
5Feed-forward Neural Networks
5.1Biological motivation
5.2Theory
5.3Learning algorithms
5.4Examples
5.5Bayesian perspectives
5.6Network complexity
5.7Approximation results
6Non-parametric Methods
6.1Non-parametric estlmation of class densities
6.2Nearest neighbour methods
6 3Learning vector quantization
6.4Mixture representations
7Tree-structured Classifiers
7.1Splitting rules
7.2Pruning rules
7.3Missing values
7.4Earlier approaches
7.5Refinements
7.6Relationships to neural networks
7.7Bayesian trees
8Belief Networks
8.1Graphical models and networks
8.2Causal networks
8 3Learning the network structure
8.4Boltzmann machines
8.5Hierarchical mixtures of experts
9Unsupervised Methods
……
10Finding Good Pattern Features
AStatistical Sidelines
Glossary
References
Author Index
Subject Index