A Model of the Acquisition and Improvement of Domain Knowledge for Functional Programming
- Publisher: IOS Press, Amsterdam
Computer science, internet | Psychology
This paper describes a model of student's knowledge growth from novice to expert within a theoretical framework of impasse-driven learning, success-driven learning and problem solving. The model represents the actual state of domain knowledge of a learner. It is designed to be part of a help system, ABSYNT, which provides user-centered help in the domain of functional programming. The model is continuously updated based on the learners programming actions. There is a distinction within the model between newly acquired and improved knowledge. Newly acquired knowledge is represented by augmenting the model with rules from the expert knowledge base. Knowledge improvement is represented by rule composition. In this way, the knowledge contained in the model is partially ordered from general rules to more specific schemas for solution fragments to specific cases (= example solutions for specific programming tasks). The model is inplemented but not yet actually used for help generation within the help system. This paper describes the theoretical framework, the ABSYNT help system, the model, a preliminary study addressing some of its empirical predictions, and the significance of the model for the help system.