Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Methodology for Building CBR Applications

: Bergmann, R.; Althoff, K.-D.


Lenz, M.; Bartsch-Spörl, B.; Burkhard, H.-D.; Wess, S.:
Case-Based Reasoning Technology. From Foundations to Applications
Berlin: Springer, 1998 (Lecture Notes in Artificial Intelligence 1400)
ISBN: 3-540-64572-1
ISSN: 0302-9743
pp.299-326 : Ill.
Book Article
Fraunhofer IESE ()
application development methodology; case-based reasoning; experience factory

The approach presented in this chapter covers the two major aspects that are important for case-based reasoning (CBR) application development to become an engineering activity. First, it presents an analytic framework for describing and classifying CBR systems and applications. This is necessary to structure the large spectrum of different systems that have already been developed. Part II of this book gives a good impression of the variety of CBR applications ranging from analytic tasks such as classification and diagnosis to synthetic tasks like design and planning. Any application developer must first analyze a new application field to decide which type of CBR approach is most appropriate. The analytic framework described in section 2 supports this analysis task. Second, the application developer must determine the specific development steps she/he has to follow in order to come to a CBR system of the desired kind. These development steps also depend very much on the particularities of the current client, like the existing organizational structure, existing IT environment, etc. Section 3 describes an experience-based approach to systematically develop a process model of how a particular CBR application should be built for a certain client. Finally, section 4 concludes by stating future directions for about how the presented CBR methodology for building and maintaining CBR applications should evolve.