Case Based Reasoning

This book presents case-based reasoning in a systematic approach with two goals: to present rigorous and formally valid structures for precise case-based reasoning, and to demonstrate the range of techniques, methods, and tools available for many applications.

Making use of different frameworks of approximate reasoning and reasoning under uncertainty, notably probabilistic and fuzzy set-based techniques, this book develops formal models of the above inference principle, which is fundamental to CBR. The case-based approximate reasoning methods thus obtained especially emphasize the heuristic nature of case-based inference and aspects of uncertainty in CBR.

Introducing issues in dynamic memory and case-based reasoning, this comprehensive volume presents extended descriptions of four major programming efforts conducted at Yale during the past several years. Each descriptive chapter is followed by a companion chapter containing the micro program version of the information. The authors emphasize that the only true way to learn and understand any AI program is to program it yourself. To this end, the book develops a deeper and richer understanding of the content through LISP programming instructions that allow the running, modification, and extension of the micro programs developed by the authors.

It also presents lessons learned about how to design CBR systems and how to apply them to real-world problems. The final chapters include a perspective on the state of the field and the most important directions for future impact.

Case-based reasoning in design is becoming an important approach to computer-support for design as well as an important component in understanding the design process. Design has become a major focus for problem solving paradigms due to its complexity and open-ended nature. This book presents a clear description of how case-based reasoning can be applied to design problems, including the representation of design cases, indexing and retrieving design cases, and the range of paradigms for adapting design cases. With a focus on design, this book differs from others that provide a generalist view of case-based reasoning. This volume provides two important contributions to the area: * a general description of the issues and alternatives in applying case-based reasoning to design, and * a description of specific implementations of case-based design. Through this combination, the reader will learn about both the general issues and the practical problems in supporting design through case-based reasoning. This book was prepared to fill a gap in the literature on the unique problems that design introduces to computational paradigms developed in computer science. It also addresses the needs of computational support for design problem solving from both theoretical and practical perspectives.

First Published in 1994. Routledge is an imprint of Taylor & Francis, an informa company.

Case-based reasoning means reasoning based on remembering previous experiences. A reasoner using old experiences (cases) might use those cases to suggest solutions to problems, to point out potential problems with a solution being computed, to interpret a new situation and make predictions about what might happen, or to create arguments justifying some conclusion. A case-based reasoner solves new problems by remembering old situations and adapting their solutions. It interprets new situations by remembering old similar situations and comparing and contrasting the new one to old ones to see where it fits best. Case-based reasoning combines reasoning with learning. It spans the whole reasoning cycle. A situation is experienced. Old situations are used to understand it. Old situations are used to solve a problem (if there is one to be solved). Then the new situation is inserted into memory alongside the cases it used for reasoning, to be used another time. The key to this reasoning method, then, is remembering. Remembering has two parts: integrating cases or experiences into memory when they happen and recalling them in appropriate situations later on. The case-based reasoning community calls this related set of issues the indexing problem. In broad terms, it means finding in memory the experience closest to a new situation. In narrower terms, it can be described as a two-part problem: assigning indexes or labels to experiences when they are put into memory that describe the situations to which they are applicable, so that they can be recalled later; and at recall time, elaborating the new situation in enough detail so that the indexes it would have if it were in the memory are identified. Case-Based Learning is an edited volume of original research comprising invited contributions by leading workers. This work has also been published as a special issues of MACHINE LEARNING, Volume 10, No. 3.