文档介绍：1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
ARTIFICIAL INTELLIGENCE
and SOFTWARE ENGINEERING
Understanding the Promise of the Future
Derek Partridge
Glenlake pany, Ltd.
Chicago • London • New Delhi
American Management Association
New York • Atlanta • Boston • Chicago • Kansas City • San Francisco • Washington, .
Brussels • Mexico City • Tokyo • Toronto
© 1998 Intellect Ltd.
ISBN: 0-8144-0441-3
All rights reserved. No part of this book may be reproduced in any form or by any means, electronic,
mechanical photocopying, recording, or otherwise without the prior written permission of the publisher.
Printed in the United States of America.
American Management Association
1601 Broadway
New York, New York 10019
Visit the American Management Association and on-line at
http:\\
Page iii
Contents
1 Introduction puter Software 1
Computers and software systems 1
An introduction to software engineering 2
Bridges and buildings versus software systems 4
The software crisis 26
A demand for more software power 29
Responsiveness to human users 29
Software systems in new types of domains 30
Responsiveness to dynamic usage environments 31
Software systems with self-maintenance capabilities 32
A need for Al systems 32
2 AI Problems and Conventional SE Problems 33
What is an AI problem? 33
Ill-defined specifications 35
Correct versus 'good enough' solutions 37
It's the HOW not the WHAT 38
The problem of dynamics 40
The quality of modular approximations 40
Context-free problems 42
3 Software Engineering Methodology 45
Specify and verify—the SAV methodology 46
The myth plete specification 47
What is verifiable? 54
Specify and test—the SAT methodology 55
Page iv
Testing for reliability 56
The strengths 57
The weaknesses 58
What are the requirements for testing? 59
What's in a specification? 61
Prototyping as a link 64
4 An Incremental and Exploratory Methodology 71
Classical methodology and A