Learning analytics;
curriculum design;
personalized study planning;
model predictive control;
receding horizon estimation;
D O I:
10.1016/j.ifacol.2022.09.225
中图分类号:
TP [自动化技术、计算机技术];
学科分类号:
0812 ;
摘要:
We propose a dynamic approach for curriculum management in university programs, i.e., for deciding which teaching and learning activities should be performed and in which order, as classes are being executed, to better aid the students reach the intended learning objectives. The approach ladders on a continuous-time dynamical model of the learning status of the individual students on the individual skills to be taught during the program. Such a model includes constructivist viewpoints on learning and zone of proximal development effects. Updating the program structure is then cast as an opportune model predictive control task, together with a moving horizon estimator that constantly infers the knowledge status of the class from the assessments performed in class. The proposed closed-loop approach is shown in simulation to significantly outperform the classical open-loop one, i.e., fixing the program structure in advance. Copyright (C) 2022 The Authors.
机构:
George Washington Univ, Grad Sch Educ & Human Dev, Washington, DC 20052 USA
209 North West St, Alexandria, VA 22314 USAGeorge Washington Univ, Grad Sch Educ & Human Dev, Washington, DC 20052 USA
Beecher, Bradley
Streitwieser, Bernhard
论文数: 0引用数: 0
h-index: 0
机构:
George Washington Univ, Grad Sch Educ & Human Dev, Washington, DC 20052 USAGeorge Washington Univ, Grad Sch Educ & Human Dev, Washington, DC 20052 USA