Knowledge for courseware engineering: A framework for inductive knowledge acquisition based upon evaluation of adaptable courseware products

Abstract

Authors or adaptors of courseware products preferably should receive support in the process of development and adaptation of courseware products. A predictive agent is defined as a system that is able to predict the expected effectiveness of various composable products from current product attributes. The described research addresses the questions of how to acquire the necessary knowledge for a predictive agent, how to organize this knowledge, and how to link it with methods and tools for courseware authoring and adaptation. We propose to use a methodology, derived from the field of machine learning, and present a framework for applying inductive knowledge acquisition based upon the empirical evaluation of adaptable courseware products.