Is CBR a Technology or a Methodology? Ian Watson AI-CBR University of Salford, Salford, M5 4WT, UK ian@ai-cbr.org www.ai-cbr.org Abstract. This paper asks whether case-based reasoning is an AI technology like rule-based reasoning, neural networks or genetic algorithms or whether it is better described as a methodology for problem solving, that may use any appropriate technology. By describing four applications of CBR, that variously use: nearest neighbour, induc- tion, fuzzy logic and SQL, the author shows that CBR is a methodology and not a technology. 1 Introduction Artificial Intelligence (AI) is often described in terms of the various technologies developed over the last three or four decades. Technologies such as logic program- ming, rule-based reasoning, neural networks, genetic algorithms, fuzzy logic, con- straint based programming and others. These technologies are characterised by spe- cific programming languages or environments (e.g., Prolog or rule-based shells) or by specific algorithms and techniques (e.g., the Rete algorithm or back propagation). Each also has, to a lesser or greater extent, laid down particular ways or methods of solving problems (e.g., A*, depth first search, generate and test) that best use the characteristics of each technology. Case-based reasoning (CBR) is a relative newcomer to AI and is commonly de- scribed as a technology like the ones listed above. This paper will show, by examin- ing four very different CBR applications, that CBR describes a methodology for problem solving but does not prescribe a specific technology. The first section of the paper briefly describes CBR and identifies what characterises a methodology in this context. The next four sections each describe an application whose authors each felt could be described as case-based reasoning. The paper then concludes with a sum- mary and discusses the implications of viewing CBR as a methodology.