Recently, the field of adaptive learning has significantly attracted researchers’ interest. Learning path adaptation problem (LPA) is one of the most challenging problems within this field. It is also a well-known combinatorial optimization problem, its main target is the knowledge resources sequencing offered to a specific learner with a specific context. The learning path candidate solutions can be only approximated as the LPA problem belongs to NP-hard problems and heuristics and meta-heuristics are usually used to solve it. In this direction, this paper summarizes existing works and presents an innovative approach modeled as an objective optimization problem, and an improved Genetic algorithm (GA) is proposed to deal with it. Our contribution does not only reduce the search space size and increase search efficiency, but it is also more explicit in finding the best composition of learning objects for a given learner. Besides the proposed GA, introduces an archive-based bag-of-operators mechanism to tackle two well-known standards GA drawbacks. The simulation results show that the proposed method makes a significant improvement compared to a well-known evolutionary approach, which is the PSO algorithm, and a random search approach. In addition, an empirical experiment is conducted and the results are very encouraging.

最近，适应性学习领域已经引起了研究者的极大兴趣。学习路径适应问题（LPA）是该领域中最具挑战性的问题之一。它也是一个众所周知的组合优化问题，其主要目标是为具有特定上下文的特定学习者提供的知识资源排序。仅当LPA问题属于NP-hard问题且通常使用启发式方法和元启发式方法来求解时，才可以近似学习候选路径的解决方案。在这个方向上，本文总结了现有工作，并提出了一种以目标优化问题为模型的创新方法，并提出了一种改进的遗传算法（GA）。我们的贡献不仅减少了搜索空间的大小并提高了搜索效率，但是在为给定的学习者找到学习对象的最佳组成方面也更加明确。除了拟议的通用航空，还引入了基于档案的操作员袋机制，以解决通用航空标准的两个弊端。仿真结果表明，与PSO算法和随机搜索方法等著名的进化方法相比，该方法具有明显的改进。另外，进行了实证实验，结果令人鼓舞。这是PSO算法，也是一种随机搜索方法。另外，进行了实证实验，结果令人鼓舞。这是PSO算法，也是一种随机搜索方法。另外，进行了实证实验，结果非常令人鼓舞。