It is difficult to generalize and accumulate experiences of system development as methodologies for building meta-learning support systems. Therefore, we need to build a framework that is useful to design and evaluate meta-learning support systems. Thus we propose a framework as a basis to design and evaluate meta-learning support systems. In this paper, we firstly describe our philosophy to solve the problem. Secondly, we propose a metalearning process model as a basis to understand meta-learning task and what kinds of factors of difficulty exist in performing meta-learning activities. Thirdly, we explain our conceptualization as a basis to design support functions for prompting meta-learning processes. Then, we integrate a meta-learning process model and the conceptualizations, so that we can design and evaluate meta-learning systems. Finally, we illustrate the usefulness of the framework by taking our presentation based meta-learning system as an example. © Springer-Verlag 2010.
CITATION STYLE
Seta, K., Fujiwara, M., Noguchi, D., Maeno, H., & Ikeda, M. (2010). Building a framework to design and evaluate meta-learning support systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6279 LNAI, pp. 163–172). https://doi.org/10.1007/978-3-642-15384-6_18
Mendeley helps you to discover research relevant for your work.