基于用户影响力感知的在线学习资源推荐方法

Online Learning Resource Recommendation Approach Based on User Influence Perception

针对在线学习中推荐数据稀疏性和资源多样化需求等问题,构建了一套在线学习混合推荐策略。该研究提出了一种基于用户相似性、知识可信度和用户影响力评估的在线学习用户模型LIAM,以提高推荐效果。同时,采用动态直觉模糊DIF策略对LIAM模型进行优化,以提高模型的推荐准确性和可解释性。最后,该研究提出了一种基于自组织的推荐方法SOR,用于解决推荐结果的多样性和覆盖性问题,从而形成了一套完整的在线学习混合推荐策略。同时,采用Coursera数据集对推荐方法SOR进行性能验证,实验结果表明,该方法优于其他两种代表性推荐方法。推荐方法SOR有望为在线学习推荐系统提供更加准确和个性化的推荐服务,提升学习效果和用户体验。

To address the issues of data sparsity in online learning and diverse resource demands,researchers have developed a hybrid recommendation strategy for online learning.This study introduces an online learning user model called LIAM,which is based on user similarity,knowledge credibility,and user influence assessment to enhance recommendation effectiveness.Additionally,the LIAM model is optimized by using a dynamic intuition fuzzy(DIF)strategy to improve recommendation accuracy and interpretability.Finally,a self-organizing recommendation method,SOR,is proposed to address the diversity and coverage of recommendation results,forming a comprehensive hybrid recommendation strategy for online learning.By using the Coursera dataset for performance validation,experimental results show that the SOR method outperforms two other representative recommendation methods.The SOR recommendation method is expected to provide more accurate and personalized recommendation services for online learning recommendation systems,thus enhancing learning outcomes and user experiences.