Data Mining and Knowledge Discovery Approaches Based on Rule Induction Techniques
This book will give the reader a perspective into the core theory and practice of data mining and knowledge discovery (DM&KD). Its chapters combine many theoretical foundations for various DM&KD methods, and they present a rich array of examples—m
- PDF / 42,822,186 Bytes
- 784 Pages / 441 x 666 pts Page_size
- 51 Downloads / 183 Views
DATA MINING AND KNOWLEDGE DISCOVERY APPROACHES BASED ON RULE INDUCTION TECHNIQUES
Edited by EVANGELOS TRIANTAPHYLLOU Louisiana State University, Baton Rouge, Louisiana, USA GIOVANNI FELICI Consiglio Nazionale delle Ricerche, Rome, Italy
^
Spri ringer
Library of Congress Control Number: 2006925174 ISBN-10: 0-387-34294-X
e-ISBN: 0-387-34296-6
ISBN-13: 978-0-387-34294-8
Printed on acid-free paper.
© 2006 Springer Science-i-Business Media, LLC All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science-fBusiness Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed in the United States of America. 987654321 springer.com
I gratefully dedicate this book to my new life's inspiration, my mother Helen and late father John (loannis), my late Grandfather (Evangelos), and also to my beloved Ragus and Ollopa ("Ikasinilab"). It would had never been prepared without their encouragement, patience, and unique inspiration. —Evangelos Triantaphyllou I wish to dedicate this book to la Didda, le Pullalle, and Misty—four special girls who are always on my side—and to all my friends, who make me strong; to them goes my gratitude for their warm support. —Giovanni Felici
TABLE OF CONTENTS List of Figures List of Tables Foreword Preface Acknowledgements
xxiii xxix xxxvii xxxix xlvii
Chapter 1 A COMMON LOGIC APPROACH TO DATA MINING AND PATTERN RECOGNITION, by A. Zakrevskij 1. Introduction 1.1 Using Decision Functions 1.2 Characteristic Features of the New Approach 2. Data and Knowledge 2.1 General Definitions 2.2 Data and Knowledge Representation the Case of Boolean Attributes 2.3 . Data and Knowledge Representation the Case of Multi-Valued Attributes 3. Data Mining - Inductive Inference 3.1 Extracting Knowledge from the Boolean Space of Attributes 3.2 The Screening Effect 3.3 Inductive Inference from Partial Data 3.4 The Case of Multi-Valued Attributes 4. Knowledge Analysis and Transformations 4.1 Testing for Consistency 4.2 Simplification 5. Pattern Recognition - Deductive Inference 5.1 Recognition in the Boolean Space 5.2 Appreciating the Asymmetry in Implicative Regularities 5.3 Deductive Inference in Finite Predicates 5.4 Pattern Recognition in the Space of Multi-Valued Attributes 6. Some Applications 7. Conclusions References Author's Biographical Statement
1 2 2 4 6 6 9 10 12 12 18 20 21 23 23 27 28 28 31 34 36 38 40 41 43
Chapter 2 THE ONE CLAUSE AT A TIME (OCAT) APPROACH TO DATA MINING AND KNOWLEDGE DISCOVERY, by
