Multi-agent model for failure assessment and diagnosis in teaching-learning processes

Abstract

Currently, there are not effective mechanisms in virtual learning environments, that allow an early detection and diagnosis of learning failures. Incorporating this kind of elements into virtual learning environments could improve learning since the diagnosis provided by the system can design an action plan that contributes to the strengthening of the virtual course topics. The aim of this paper is to present the design and development of a multi-agent model for the assessment and diagnosis of failures which seeks to discover the shortcomings in learning from the virtual assessment process. In addition, the model looks for offering feedback and recommending new educational resources adapted to the learner’s profile. Based on the proposed model, a prototype was implemented and validated through a case study. The results obtained allow us to conclude that the students felt accompanied during the assessment process and obtained real-time feedback that identified shortcomings and allowed to recommend educational resources in order to strengthen their learning process.