Probabilistic Learner Modeling in Scientific Inquiry Exploratory Learning Environment

Research on learning has shown that although computer-based exploratory learning environments have been proven to be beneficial to learners, effectively inferring a learner's actions under a sound teaching and learning model that enhances exploratory behaviours remains uncertain. To address this problem, this article aims at discussing and highlighting the detail methodological approach for designing, and integrating the probabilistic learner modeling leveraging Bayesian networks into Scientific Inquiry Exploratory Learning Model. This integration mainly serves as a basis to support learners with adaptive instructions and facilitating the acquisition of both domain knowledge as well as scientific inquiry skills. To visualize the proposed methodological approach, a computer-based scientific inquiry exploratory learning environment named InQPro is developed. This article ends with presenting the preliminary investigation on employing Artificial Students technique to investigate the propagation of probabilities between subnetworks, and identifying threshold parameters in the probabilistic learner model.