Combining nominal and continuous properties in an incremental learning system for design

Abstract: "Research in machine learning has produced many learning algorithms mainly for classification tasks. This paper reports on an extension made to the learning program COBWEB to allow it to handle examples described by a more complex description language. The paper describes Bridger, a system that implements this, as well as other extensions. Bridger, combines learning and performance in design tasks. The extension implemented has been tested succes[s]fully in four design domains. This extension and others are necessary for allowing Bridger to master design rather than simple classification tasks."