Skip to main content

1997 | Buch

Reasoning with Complex Cases

verfasst von: Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz

Verlag: Springer US

Buchreihe : The International Series in Engineering and Computer Science

insite
SUCHEN

Über dieses Buch

Reasoning with Complex Cases emphasizes case retrieval methods based on structured cases as they are relevant for planning, configuration, and design, and provides a systematic view of the case reuse phase, centering on complex situations. So far, books on case-based reasoning considered comparatively simple situations only. This book is a coherent work, not a selection of separate contributions, and consists largely of original research results using examples taken from industrial design, biology, medicine, jurisprudence and other areas.
Reasoning with Complex Cases is suitable as a secondary text for graduate-level courses on case-based reasoning and as a reference for practitioners applying conventional CBR systems or techniques.

Inhaltsverzeichnis

Frontmatter

Reasoning with Cases

Frontmatter
1. Introduction
Abstract
Do you remember the strawberry cake that was so well received at last month’s party? Well, the strawberry season is over, but replacing them by raspberries might be a good idea.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
2. Reasoning with Complex Cases
Abstract
Confronted with a problem, humans often remember previous problem solving episodes or solutions and try to modify them for their current situation (see figure 2.1). Case-based reasoning is an attempt to support this approach with a computer. A quick tour through case-based reasoning in section 2.1 gives a first impression of the main concepts and some central problems encountered in this field. All the notions and procedures encountered there will be reconsidered in more detail in sections 2.2 to 2.8. The chapter closes with a summary characterizing complex cases.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
3. The FABEL Domain for Illustration
Abstract
In this book we will give a lot of examples to illustrate the CBR issues and methods discussed. Many of the examples given come from the FABEL project and its application domain, that is architectural design. In this chapter we introduce this domain.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz

Case Retrieval

Frontmatter
4. Case Similarities
Abstract
As the name implies, case-based reasoning uses previous, known cases to solve a new problem. Sometimes it is easy to find an appropriate case, but usually this is no trivial task. In this chapter we will explicate the notion of case similarity and the reason for using it in case retrieval.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
5. Similarities Between Attributes
Abstract
Retrieval systems have been used for decades and brought to some perfection for document retrieval. No wonder they are being employed for cases, too, even probably in the majority of applications.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
6. Restricted Geometric Relationships
Abstract
It is hard to capture geometric relationships by means of attributes and attribute values; it is even harder and in most cases practically impossible to base retrieval on such a description since the names of the involved objects are irrelevant and mostly unknown; what matters are the relationships between corresponding objects in the query and source cases. Therefore one has to develop specialized retrieval methods.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
7. Graphs
Abstract
Sometimes cases are represented as graphs of various brands — directed or undirected, labeled or not, belonging to special subclasses like planar or connected graphs. There exist already many graph matching procedures outside case-based reasoning. Some of them have been adapted to purposes of case retrieval.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
8. Semantic Nets
Abstract
Semantic nets have been used in artificial intelligence for a long time in various ways, e.g. for reasoning. Some types of semantic nets use a method known as activation spreading for finding subnets that are semantically related to some initially activated nodes. If in addition certain subnets can be identified as cases, this can be considered a special form of case retrieval.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
9. Model-Based Similarities
Abstract
The case retrieval systems in the preceding chapters employ more or less domain-independent methods. Some other systems utilize deep knowledge of the special application; they are based on some kind of domain model. Retrieval tends to be more complicated than just comparing two cases; the notion of similarity becomes blurred.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
10. Enhancing The Quality of Retrieval Methods
Abstract
In the preceding chapters we have analyzed several methods for case retrieval. These form the essential part in the retrieval process. It is however worthwhile to review procedures that enhance the quality of the pure retrieval algorithms. This is the purpose of the present chapter.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
11. Conclusions and Perspectives
Abstract
Having explored a wealth of existing case retrieval systems and their underlying methods, we return to the role of similarity as a substitute for the usefulness of a case, discuss the problem of assessing retrieval quality and connect retrieval to the next step in case-based reasoning, reuse.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz

Case Reuse

Frontmatter
12. Reusing Cases
Abstract
Having retrieved one or more useful cases, one reuses them for solving the present problem. The variety of reuse procedures and methods can be analyzed from several perspectives: application type (analysis or synthesis); technical aspects; steps in the reuse process. An overview is given in this chapter; some issues are expanded in the following ones.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
13. Reuse of a Single Case: Adaptation
Abstract
This chapter proposes a framework for describing systems that reuse a single source case. It applies it to the systems selected and extracts a first set of guidelines. The major issue in the design of such systems is the approach to adaptation, because it influences many other decisions. Adaptation remains an issue with case-based reasoning systems that reuse multiple source cases. But there, strategic questions of how to handle more than one source case move into foreground. Therefore, this chapter focuses on case adaptation and the next one on strategies.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
14. Reuse of Multiple Cases: Strategies
Abstract
In the previous chapter it was assumed that the given problem can be interpreted as a query case and that a single source case is sufficient to solve the problem. This assumption will now be relaxed. As in the previous chapter, a framework will be proposed, it will be applied to the systems, and some guidelines will be extracted.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
15. Source Cases — The Third Factor
Abstract
The conclusions from the micro level of single case adaptation in section 13.3 and from the macro level of multiple case reuse in section 14.3 indicate that both aspects are related to a third factor in the design of a case reuse system: the size and content of the source cases and the relations between them. This chapter is devoted to this still neglected yet central issue. The systems need not be revisited, because source cases have been discussed together with the strategies in section 14.2. A last group of guidelines will be given to relate the choice of source cases to the approach to adaptation and to the strategy.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
16. Conclusions and Perspectives
Abstract
Looking back to the preceding chapters, we see that cases as well as reuse methods are more complex than usually anticipated. Our analysis leads to new insights of case-based reasoning and to a new assessment of potentials and risks.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz

An Integrated System

Frontmatter
17. Designing Integrated CBR Systems
Abstract
Most existing CBR systems focus on one particular issue, e.g. a retrieval or a reuse method. Therefore no unusual questions of system architecture arise. If however many methods are to be combined, this will lead to problems. FABEL PROTOTYPE is the only known CBR system combining a bunch of quite diverse methods. Therefore, this part of the book on integrated systems is predominantly devoted to FABEL PROTOTYPE. In the present chapter, we deal with questions of knowledge acquisition for complex CBR systems and with general problems of CBR architectures and we illustrate the FABEL PROTOTYPE solution in a non-technical overview.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
18. Software Architecture
Abstract
A lot of approaches to case retrieval and case reuse were presented in parts II and III of this book. Some of them are integrated in FABEL research Prototype. It contains the eight problem solving tools AAAO, ANOPLA, ASM, ASPECT, CHECKUP, RABIT, SYN*, and TOPO, which support altogether the CBR steps retrieval and reuse; two of them (AAAO and ANOPLA) also support the construction of new solutions from scratch. Additionally, there are two tools, CBASE and CSET, for maintaining source cases, case bases, and case sets.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
19. Scenario of Application
Abstract
In this last chapter we want to become very concrete about the use of FABEL PROTOTYPE.
Friedrich Gebhardt, Angi Voß, Wolfgang Gräther, Barbara Schmidt-Belz
Backmatter
Metadaten
Titel
Reasoning with Complex Cases
verfasst von
Friedrich Gebhardt
Angi Voß
Wolfgang Gräther
Barbara Schmidt-Belz
Copyright-Jahr
1997
Verlag
Springer US
Electronic ISBN
978-1-4615-6233-7
Print ISBN
978-1-4613-7859-4
DOI
https://doi.org/10.1007/978-1-4615-6233-7