Structural Similarity as Guidance in Case-Based Design

This paper presents a novel approach to determine structural similarity as guidance for adaptation in case-based reasoning (Cbr). We advance structural similarity assessment which provides not only a single numeric value but the most speci c structure two cases have in common, inclusive of the modi cation rules needed to obtain this structure from the two cases. Our approach treats retrieval, matching and adaptation as a group of dependent processes. This guarantees the retrieval and matching of not only similar but adaptable cases. Both together enlarge the overall problem solving performance of Cbr and the explainability of case selection and adaptation considerably. Although our approach is more theoretical in nature and not restricted to a speci c domain, we will give an example taken from the domain of industrial building design. Additionally, we will sketch two prototypical implementations of this approach.

• Type: Conference
• Year: 1993
• Where: EWCBR
• Authors: Katy Börner
• Keywords: Approach Treats Retrieval, Automated Reasoning, EWCBR 1993, Structural Similarity, Structural Similarity Assessment