Vania Bessa Machado and Bert Bredeweg
Department of Social Science Informatics (SWI), University of Amsterdam, Roetersstraat 15, 1018 WB Amsterdam, The Netherlands Email: {vania, bert}@swi.psy.uva.nl
Submitted to the 10th International Conference on Artificial Intelligence in Education AI-ED 2001, May 19-23, San Antonio, Texas, USA.
Keywords: Qualitative Simulation, Model-based Reasoning, Model Construction, Learning by Building Models
Abstract: Having learners construct computer-based simulations is becoming increasingly important as an approach to induce learning. Qualitative simulations incorporate a rich vocabulary for articulating insights about systems and their behaviours, including notions such as structural constituents, identifying qualitatively distinct behaviours, and making explicit the causal dependencies that govern a system's behaviour. When building a qualitative model all these details have to be made explicit (they have to be represented in the model). As a result, building a qualitative model induces a deep understanding of the system and its behaviour. In our research we want to exploit this phenomenon, i.e. have learners learn by building qualitative simulations. However, building qualitative models is generally seen as a difficult task, and no easy-to-use tools exist to support a learner in performing such a task. In this paper, we present MOBUM, a domain independent model building environment aimed at supporting learners in building qualitative simulations. MOBUM is a fully implemented prototype in JAVA. The main goal of this prototype is to generate and clarify ideas with respect to how such model building tool should be constructed.
Full Paper (PDF 210 KB)