Case-based reasoning relies on the hypothesis that “similar problems have similar solutions,” which seems to apply, in a certain sense, to a large range of applications. In order to be generally applicable and useful for problem solving, however, this hypothesis and the corresponding process of case-based inference have to be formalized adequately. This paper provides a formalization which makes the “similarity structure” of a system accessible for reasoning and problem solving. A corresponding (constraint-based) approach to case-based inference exploits this structure in a way which allows for deriving a similarity-based prediction of the solution to a target problem in form of a set of possible candidates (supplemented with a level of confidence.)
CITATION STYLE
Hüllermeier, E. (1999). Exploiting similarity for supporting data analysis and problem solving. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1642, pp. 257–268). Springer Verlag. https://doi.org/10.1007/3-540-48412-4_22
Mendeley helps you to discover research relevant for your work.