Elaborating Analogies from Conceptual Models George Spanoudakis Department of Computer Science, The City University, U.K Email: gespan@cs.city.ac.uk Panos Constantopoulos Department of Computer Science, University of Crete, Greece and Institute of Computer Science, Foundation for Research and Technology-Hellas Email: panos@ics.forth.gr Abstract. This paper defines and analyses a computational model of similarity which detects analogies between objects based on conceptual descriptions of them, constructed from classification, generalization relations and attributes. Analogies are detected(elaborated) by functions which measure conceptual distances between objects with respect to these semantic modelling abstractions. The model is domain independent and operational upon objects described in non uniform ways. It doesn’t require any special forms of knowledge for identifying analogies and distinguishes the importance of distinct object elements. Also, it has a polynomial complexity. Due to these characteristics, it may be used in complex tasks involving intra or inter-domain analogical reasoning. So far the similarity model has been applied in the domain of software engineering. First, to support the specification of software requirements by analogical reuse and second, to enable the integration of requirements specifications, generated by the multiple agents involved in information system development. Details of these applications can be found in sited references. Also, we have conducted an empirical evaluation of: (i) the consistency of the estimates generated by the model against human intuition about similarity and (ii) its recall performance in tasks of analogi- cal retrieval, the results of which are presented in this paper. 1. Introduction The remarkable ability of humans to understand novel situations by analogy to familiar ones and to solve new problems by remembering solutions to old analogous ones, has motivated the study of anal- ogy as a non deductive paradigm of reasoning and its application in a variety of domains including law[4], medicine[16], economics[3], mechanical[59]) and software engineering[44,53,54,64,75,76,77]. Analogy has been studied in cognitive science, psychology, philosophy and artificial intelligence, yet from different perspectives. Cognitive scientists and psychologists are primarily concerned with how humans recall analogs and reason by analogy[22,24,47,56]. Philosophers concentrate on prerequisite conditions for drawing valid analogical inferences[38,46]. Finally, AI researchers focus on the development of computational models and systems for reasoning by analogy[26,31]. Along the latter direction of research, most of the computational models of analogical reasoning were developed as models of human cognition[20,28,47,65] without emphasizing efficiency and usability aspects, during the eighties. As a result, these models were very sensitive to the representation, the