Argo: a system for design by analogy

The authors have developed domain-independent techniques for reasoning and learning by analogy, techniques that capture and apply problem-solving experience and that are useful for search-intensive problems found in the design domain. They have incorporated these techniques, plus nonmonotonic reasoning capability, into Argo, a tool for building knowledge-based systems that improve with use. They discuss learning from experience and also analogical reasoning and learning. They describe knowledge representation and interference in Argo and Argo´s control strategy for design. An application to VLSI circuit design is provided. Various features of Argo are described.<>