文档介绍：Introduction
to
Machine
Learning
Ethem Alpaydm
The MIT Press
Cambridge, Massachusetts
London, England
© 2004 Massachusetts Institute of Technology
All rights reserved. No part of this book may be reproduced in any form by any
electronic or mechanical means (including photocopying, recording, or informa•
tion storage and retrieval) without permission in writing from the publisher.
MIT Press books may be purchased at special quantity discounts for business
or sales promotional use. For information, please email ******@mitpress.
or write to Special Sales Department, The MIT Press, 5 Cambridge Cen•
ter, Cambridge, MA 02142.
Ubrary of Congress Control Number: 2004109627
ISBN: 0-262-01211-1 (hc)
Typeset in 10/13 Lucida Bright by the author using ~TEX 2E.
Printed and bound in the United States of America.
**********
Contents
Series Foreword xiii
Figures xv
Tables xxiii
Preface xxv
Acknowledgments xxvii
Notations xxix
1 Introduction 1
What Is Machine Learning? 1
Examples of Machine Learning Applications 3
Learning Associations 3
Classification 4
Regression 8
Unsupervised Learning 10
Reinforcement Learning 11
Notes 12
Relevant Resources 14
Exercises 15
References 16
2 Supervised Learning 17
Learning a Class from Examples 17
Vapnik-Chervonenkis (VC) Dimension 22
Probably Approximately Correct (PAC) Learning 24
vi Contents
Noise 25
Learning Multiple Classes 27
Regression 29
Model Selection and Generalization 32
Dimensions of a Supervised Machine Learning Algorithm 35
Notes 36
Exercises 37
References 38
3 Bayesian Decision Theory 39
Introduction 39
Classification 41
Losses and Risks 43
Discriminant Functions 45
Utility Theory 46
Value of Information 47
works 48
Influence Diagrams 55
Association Rules 56
Notes 57