Quinlan C4.5

Programs for Machine Learning
1. Auflage 2014
ISBN: 978-0-08-050058-4
Verlag: Elsevier Science & Techn.
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 312 Seiten, Web PDF

Classifier systems play a major role in machine learning and knowledge-based systems, and Ross Quinlan's work on ID3 and C4.5 is widely acknowledged to have made some of the most significant contributions to their development. This book is a complete guide to the C4.5 system as implemented in C for the UNIX environment. It contains a comprehensive guide to the system's use , the source code (about 8,800 lines), and implementation notes. C4.5 starts with large sets of cases belonging to known classes. The cases, described by any mixture of nominal and numeric properties, are scrutinized for patterns that allow the classes to be reliably discriminated. These patterns are then expressed as models, in the form of decision trees or sets of if-then rules, that can be used to classify new cases, with emphasis on making the models understandable as well as accurate. The system has been applied successfully to tasks involving tens of thousands of cases described by hundreds of properties. The book starts from simple core learning methods and shows how they can be elaborated and extended to deal with typical problems such as missing data and over hitting. Advantages and disadvantages of the C4.5 approach are discussed and illustrated with several case studies. This book should be of interest to developers of classification-based intelligent systems and to students in machine learning and expert systems courses.

J. Ross Quinlan, University of New South Wales

Autoren/Hrsg.

Quinlan, J. Ross

Inhaltsverzeichnis

1;Front Cover;1
2;C4.5: Programs for Machine Learning;4
3;Copyright Page;5
4;Table of Contents;6
5;Preface;8
6;Obtaining the C4.5 Code;10
7;CHAPTER 1. Introduction;12
7.1;1.1 Example: Labor negotiation settlements;14
7.2;1.2 Other kinds of classification models;23
7.3;1.3 What lies ahead;27
8;CHAPTER 2. Constructing Decision Trees;28
8.1;2.1 Divide and conquer;28
8.2;2.2 Evaluating tests;31
8.3;2.3 Possible tests considered;35
8.4;2.4 Tests on continuous attributes;36
9;CHAPTER 3. Unknown Attribute Values;38
9.1;3.1 Adapting the previous algorithms;39
9.2;3.2 Play/Don't Play example again;41
9.3;3.3 Recapitulation;43
10;CHAPTER 4. Pruning Decision Trees;46
10.1;4.1 When to simplify?;47
10.2;4.2 Error-based pruning;48
10.3;4.3 Example: Democrats and Republicans;52
10.4;4.4 Estimating error rates for trees;53
11;CHAPTER 5. From Trees to Rules;56
11.1;5.1 Generalizing single rules;58
11.2;5.2 Class rulesets;61
11.3;5.3 Ranking classes and choosing a default;65
11.4;5.4 Summary;66
12;CHAPTER 6. Windowing;68
12.1;6.1 Example: Hypothyroid conditions revisited;69
12.2;6.2 Why retain windowing?;69
12.3;6.3 Example: The multiplexor;71
13;CHAPTER 7. Grouping Attribute Values;74
13.1;7.1 Finding value groups by merging;75
13.2;7.2 Example: Soybean diseases;76
13.3;7.3 When to form groups?;77
13.4;7.4 Example: The Monk's problems;78
13.5;7.5 Uneasy reflections;80
14;CHAPTER 8. Interacting with Classification Models;82
14.1;8.1 Decision tree models;82
14.2;8.2 Production rule models;89
14.3;8.3 Caveat;91
15;CHAPTER 9. Guide to Using the System;92
15.1;9.1 Files;92
15.2;9.2 Running the programs;95
15.3;9.3 Conducting experiments;100
15.4;9.4 Using options: A credit approval example;102
16;CHAPTER 10. Limitations;106
16.1;10.1 Geometric interpretation;106
16.2;10.2 Nonrectangular regions;107
16.3;10.3 Poorly delineated regions;109
16.4;10.4 Fragmented regions;111
16.5;10.5 A more cheerful note;113
17;CHAPTER 11. Desirable Additions;114
17.1;11.1 Continuous classes;114
17.2;11.2 Ordered discrete attributes;115
17.3;11.3 Structured attributes;115
17.4;11.4 Structured induction;116
17.5;11.5 Incremental induction;117
17.6;11.6 Prospectus;118
18;Appendix: Program Listings;120
18.1;Brief descriptions of the contents of the files;121
18.2;Notes on some important data structures;123
18.3;File Makefile;126
18.4;Alphabetic index of routines;299
18.5;References and Bibliography;302
19;Author Index;308
20;Subject Index;310