针对目前启发式算法用于解决个性化网络学习资源推荐问题时存在推荐速度较慢、不稳定等问题,文章提出基于改进二进制粒子群算法的个性化网络学习资源推荐方法(AsyBPSO-RA).该方法将个性化网络学习资源推荐问题建构为适应度函数,利用改进二进制粒子群算法(AsyBPSO)优化此适应度函数,生成推荐结果;AsyBPSO采用非对称映射函数,取代基本二进制粒子群算法中的S型映射函数,以更好地平衡算法的探索和开发阶段.通过五组实验结果对比分析发现,AsyBPSO收敛能力强,稳定性高,表明AsyBPSO-RA是较为有效的个性化网络学习资源推荐方法. In order to solve the problem of slow recommendation and instability of current heuristic algorithm to solve the problem of personalized online learning resource recommendation, this paper proposes AsyBPSO-RA based on improved binary particle swarm optimization algorithm. The method constructs the personalized online learning resource recommendation problem as a fitness function, and optimizes the fitness function by using improved binary particle swarm optimization (AsyBPSO) to generate the recommended result. AsyBPSO uses the asymmetric mapping function instead of the basic binary particle swarm algorithm Type mapping function to better balance the exploration and development stage of the algorithm.Comparison of the five experimental results shows that AsyBPSO has strong convergence ability and high stability, which indicates that AsyBPSO-RA is a more effective method to recommend personalized online learning resources.