The "what" and "how" of learning in design

Abstract

Previous experiences hold a wealth of knowledge which we often take for granted and use unknowingly through our every day working lives. In design, those experiences can play a crucial role in the success or failure of a designproject, having a great deal of influence on the quality, cost and development time of a product. But how can we empower computer based design systems to acquire this knowledge? How would we use such systems to support design? This paper outlines some of the work which has been carried out in applying and developing Machine Learning techniques to support the design activity; particularly in utilising previous designs and learning the design process