Student Modelling and Knowledge Acquisition Process in Complex Nature Domains

• Published in: Higher Education for All. From Challenges to Novel Technology-Enhanced Solutions Vol. 832; pp. 89 - 106
• Main Authors: Maschio, Eleandro, Moreira, Carolina
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2018; Springer International Publishing
• Series: Communications in Computer and Information Science
• ISBN: 9783319979335; 3319979337
• ISSN: 1865-0929; 1865-0937
• DOI: 10.1007/978-3-319-97934-2_6

This chapter discusses how high-level knowledge about human expertise features can be modelled, represented and further interpreted to support learning and tutoring interactions, promoting adaptivity. For this, as an instance, the research focuses on the domain of computer programming. In the tutoring of many domains of complex nature, most past work has tended to concentrate on the theoretical principles of how humans acquire expertise. The few implementations there have been are domain-specific. However, the problem of providing an epistemology for describing knowledge about problem statements and students’ states of belief has been neglected. This research treats these problems through (1) a method based on genetic graphs for managing the complexity of courseware authoring of problem statements and (2) a general process for dynamic modelling a learner’s knowledge by overlaying it against the domain expertise features. Both the method and the model are supported by implemented prototype software tools that integrate the educational environment MULTIPLA. To evaluate them, empirical observations have been carried out, focusing on the generality of genetic graphs as an authoring language to provide a unified expert-student framework for developing skills.