Abstract:

Single-purpose learning algorithms are often not efficient in acquiring knowledge for expert systems in design. This paper presents a method that learns design knowledge from design cases by the extensive generalisations of abstract design schemas which function as both a representation schema storing generalised knowledge and a memory organisation facility indexing specific cases and aiding in retrieval. Unlike most current machine learning algorithms, it aims at multiple targets of prediction, feature reminding, case retrival and new case generation in a single system.