基于多维特征差异的个性化学习资源推荐方法

发布时间：2018-07-22 16:22

【摘要】：协同过滤算法和二进制粒子群算法是目前学习资源推荐领域研究热点.然而,协同过滤算法推荐的学习资源过于随机化,不能满足学习者进行整体知识建构的要求.而基于二进制粒子群算法构建的资源推荐模型,以推荐所有学习者完整的学习资源为目标,且模型数据较难预测,不符合在线智能化学习的趋势.针对以上问题,提出了基于多维特征差异的个性化学习资源推荐算法:首先根据学习者和学习资源多维特征差异建立学习资源推荐模型,并考虑了学习偏好;其次引入协同过滤技术对模型数据进行预测;最后针对推荐模型的多目标优化特征,将协同过滤算法和二进制粒子群算法结合,提出了对惯性权重和种群多样性进行动态协同调整的自适应二进制粒子群算法,实现了个性化学习资源推荐.实验证明,该算法具有较好的准确性,能够满足个性化学习资源推荐的需要.
[Abstract]:Collaborative filtering algorithm and binary particle swarm optimization (BPSO) are the research focus in the field of learning resource recommendation. However, the learning resources recommended by collaborative filtering algorithm are too randomized to meet the requirements of learners' overall knowledge construction. The resource recommendation model based on binary particle swarm optimization algorithm aims at recommending all learners' complete learning resources, and the model data is difficult to predict, which is not in line with the trend of online intelligent learning. Aiming at the above problems, a personalized learning resource recommendation algorithm based on multidimensional feature difference is proposed: firstly, the learning resource recommendation model is established according to the multi-dimensional characteristic difference of learner and learning resource, and learning preference is considered. Secondly, the collaborative filtering technology is introduced to predict the model data. Finally, the cooperative filtering algorithm and binary particle swarm optimization algorithm are combined to solve the multi-objective optimization characteristics of the recommendation model. An adaptive binary particle swarm optimization (BPSO) algorithm for dynamic coordination of inertial weight and population diversity is proposed and personalized learning resource recommendation is realized. Experiments show that the algorithm has good accuracy and can meet the needs of personalized learning resources recommendation.
【作者单位】：浙江工业大学教育科学与技术学院;浙江工业大学计算机科学与技术学院;
【基金】：国家自然科学基金(61503340) 国家社会科学基金(16BTQ084)~~
【分类号】：TP391.3

【相似文献】