Individualizing HCI in e-learning through assessment approach

• Autores: Brahim El Falaki, Nour-Eddine El Faddouli, Mohammed Khalidi Idrissi, Samir Bennani
• Localización: The International journal of engineering education, ISSN-e 0949-149X, Vol. 29, no. Extra 3, 2013, págs. 650-659
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • In e-learning systems, learners can have the same goal, but not necessarily the same predisposition, or the same knowledge.Therefore, an optimal pedagogical path for one does not necessarily fit the other. Thus, providing an interactive environmenttailored to the learner’s needs is one of the most important goals of e-learning environments. In our proposal, we make theadaptation of learning our main objective. Our theoretical framework stems from the fact that HCI is a combination of cognitive,behavioral and computer sciences. On the cognitive and behavioural scope, we have opted for adaptive formative assessment so asto identify the learner’s competence level and, thereon, to guide the learner to reach the educationally drawn output profile. Theaforementioned assessment highlights the learner’s real time competence. The latter and the learner’s prior knowledge are pivotalelements in adapting the learning process. On the computing scope, this process is carried out via an e-learning system in which theproposed assessment is implemented through Services Oriented Architecture (SOA).In this paper, it would be prominent to individualize the learning path by adopting formative assessment by proposing anadaptive test which offers a selection of optimal items in a sequence taking into account the profile and the progress of the learner.To implement the proposed system, first, we modelled learner and items according to competency based approach (CBA). Then,modelled the formative assessment in an adaptive approach using the Item response theory (IRT), this will provide a series ofconsecutively selected items. The answer to an item determines the selection of the next one taking into account the previousresponses and performances recorded in the learner model.