Solving the Adaptive Curriculum Sequencing Problem with Prey-Predator Algorithm

Abstract

Adaptive curriculum sequencing (ACS) is still a challenge in the adaptive learning field. ACS is a NP-hard problem especially considering the several constraints of the student and the learning material when selecting a sequence from repositories where several sequences could be chosen. Therefore, this has stimulated several researchers to use evolutionary approaches in the search for satisfactory solutions. This work explores the use of an adaptation of the prey-predator algorithm for the ACS problem. Pedagogical experiments with a real student dataset and convergence experiments with a synthetic dataset have shown that the proposed solution is suitable for the problem, although it is a solution not yet explored in the adaptive learning literature.

A comparative analysis of metaheuristics applied to adaptive curriculum sequencing

“…Several works evaluate their proposed solution by intrinsic evaluations, i.e., checking if the chosen metaheuristic can converge to coherent solutions, observing fitness values (Machado et al. 2019 ; de Marcos et al. 2011 ; Chu et al.…”

“…It is considered a crucial issue in personalized learning as its purpose is to find the best sequence of learning materials that meets the student profile (Niknam and Thulasiraman 2020 ; Machado et al. 2019 ; Muhammad et al. 2016 ).…”

“…For instance, a Prey–Predator Algorithm (PPA) was proposed in (Machado et al. 2019 ) to address the ACS problem assuming that such metaheuristics outperform GA and PSO (two commonly metaheuristics used for ACS problem) in benchmark problems. However, the authors focused on pedagogical results and no comparison among the approaches were performed using the problem depicted here.…”

Geração de Sequências Curriculares Adaptativas baseada no Perfil dos Alunos e Materiais Didáticos utilizando o Algoritmo NSGA-III

“…O estudo realizado por [Machado et al 2019] levantou os parâmetros de materiais e alunos mais comuns nos trabalhos de SCA. Ainda, verificou quais os objetivos mais utilizados na literatura, chegando a um total de cinco objetivos.…”

Evaluating Educational Recommendation Systems: a systematic mapping

“…According to [Erdt et al 2015], this term is used to describe a technological application in teaching and learning. The TEL approach has been used for different types of recommendations, such as recommending materials [De Medio et al 2020, Machado et al 2019, exercises [Lv et al 2018], learning paths [Machado et al 2020, Nabizadeh et al 2020, partners for study groups [Khosravifar et al 2018], etc. For these works, classic techniques used in recommendation systems are used, such as content recommendation, collaborative filtering, and others.…”