Lan H.Witten,新西蘭懷卡托大學計算機科學系教授。他是ACM和新西蘭皇家學會的成員,并參加了英國、美國、加拿大和新西蘭的專業(yè)計算、信息檢索、工程等協(xié)會。他著有多部著作,是多家技術(shù)雜志的作者,發(fā)表過大量論文。
圖書目錄
Foreword vii Preface xvii 1 What's it all about? 1.1 Data mining and machine learning Describing structural patterns Machine learning Data mining 1.2 Simple examples: The weather problem and others The weather problem Contact lenses: An idealized problem Irises: A classic numeric dataset CPU performance: Introducing numeric prediction Labor negotiations: A more realistic example Soybean classification: A classic machine learning success 1.3 Fielded applications Decisions involving judgment Screening images Load forecasting Diagnosis Marketing and sales 1.4 Machine learning and statistics 1.5 Generalization as search Enumerating the concept space Bias 1.6 Data mining and ethics 1.7 Further reading 2 Input Concepts, instances, attributes 2.1 What's aconcept? 2.2 What's in an example? 2.3 What's in an attribute? 2.4 Preparing the input Gathering the data together Arff format Attribute types Missing values Inaccurate values Getting to know your data 2.5 Further reading 3 Output: Knowledge representation 3.1 Decision tables 3.2 Decision trees 3.3 Classification rules 3.4 Association rules 3.5 Ruleswith exceptions 3.6 Rules involving relations 3.7 Trees for numeric prediction 3.8 Instance-based representation 3.9 Clusters 3.10 Further reading 4 Algorithms: The basic methods 4.1 Inferring rudimentary rules Missing values and numeric attributes Discussion 4.2 Statistical modeling Missing values and numeric attributes Discussion 4.3 Divide and conquer: Constructing decision trees Calculating information Highly branching attributes Discussion 4.4 Covering algorithms: Constructing rules Rules versus trees A simple covering algorithm Rules versus decision lists 4.5 Mining association rules Item sets Association rules Generating rules efficiently Discussion 4.6 Linear models Numeric prediction Classification Discussion 4.7 Instance-based learning The distance function Discussion 4.8 Further reading 5 Credibility: Evaluating what's been learned 5.1 Training and testing 5.2 Predicting performance 5.3 Cross-validation 5.4 Other estimates Leave-one-out The bootstrap 5.5 Comparing data mining schemes 5.6 Predicting probabilities Quadratic loss function Informational loss function Discussion 5.7 Counting the cost Lift charts ROC curves Cost-sensitive learning Discussion 5.8 Evaluating numeric prediction 5.9 The minimum description length principle 5.10 Applying MDL to clustering 5.11 Further reading 6 Implementations: Real machine learning schemes 6.1 Decision trees Numeric attributes Missing values Pruning Estimating error rates Complexity of decision tree induction From trees to rules C4.5: Choices and options Discussion 6.2 Classification rules Criteria for choosing tests Missing values, numeric attributes Good rules and bad rules Generating good rules Generating good decision lists Probability measure for rule evaluation Evaluating rules using a test set Obtaining rules from partial decision trees Rules with exceptions Discussion 6.3 Extending linear classification: Support vector machines The maximum margin hyperplane Nonlinear class boundaries Discussion 6.4 Instance-based learning Reducing the number of exemplars Pruning noisy exemplars Weighting attributes Generalizing exemplars Distance functions for generalized exemplars Generalized distance functions Discussion 6.5 Numeric prediction Model trees Building the tree Pruning the tree Nominal attributes Missing values Pseudo-code for model tree induction Locally weighted linear regression Discussion 6.6 Clustering Iterative distance-based clustering Incremental clustering Category utility Probability-based clustering The EM algorithm Extending the mixture model Bayesian clustering Discussion 7 Moving on: Engineering the input and output 7.1 Attribute selection Scheme-independent selection Searching the attribute space Scheme-specific selection 7.2 Discretizing numeric attributes Unsupervised discretization Entropy-based discretization Other discretization methods Entropy-based versus error-based discretization Converting discrete to numeric attributes 7.3 Automatic data cleansing Improving decision trees Robust regression Detecting anomalies 7.4 Combining multiple models Bagging Boosting Stacking 258 Error-correcting output codes 7.5 Further reading 8 Nuts and bolts: Machine learning algorithms in Java 8.1 Getting started 8.2 Javadoc and the class library Classes, instances, and packages The weka. core package The weka. classifiers package Other packages Indexes 8.3 Processing datasets using the machine learning programs Using M5' Generic options Scheme-specific options Classifiers Meta-learning schemes Filters Association rules Clustering 8.4 Embedded machine learning A simple message classifier 8.5 Writing new learning schemes An example classifier Conventions for implementing classifiers Writing filters An example filter Conventions for writing filters 9 Looking forward 9.1 Learning from massive datasets 9.2 Visualizing machine learning Visualizing the input Visualizing the output 9.3 Incorporating domain knowledge 9.4 Text mining Finding key phrases for documents Finding information in running text Soft parsing 9.5 Mining the World Wide Web 9.6 Further reading References Index About the authors