摘要
提出了一种融合知识图谱与深度学习的优化策略,以提升个性化学习推荐系统的推荐质量和用户体验,解决数据稀疏性与冷启动问题。通过构建涵盖学习者、教育资源及其关联关系的知识图谱,为系统提供语义信息和结构化知识支持,并利用余弦相似度匹配学习者属性与资源特征,精准满足需求。同时,结合深度神经网络捕捉偏好与资源特性间的非线性关联,进一步优化推荐性能。实验结果表明,该方法在预测准确性、推荐覆盖率和多样性上分别提升11.8%、14.3%和9.7%,并在用户满意度与知识掌握度等指标上表现出优势。
This study proposes an optimization strategy integrating knowledge graphs and deep learning to enhance the recommendation quality and user experience of personalized learning recommendation systems,addressing data sparsity and cold start issues.By constructing a knowledge graph encompassing learners,educational resources,and their interrelationships,the system gains semantic information and structured knowledge support.Cosine similarity is employed to match learner attributes with resource features,thereby precisely meeting demands.Concurrently,deep neural networks capture nonlinear associations between preferences and resource characteristics to further optimize recommendation performance.Experimental results demonstrate that this method achieves improvements of 11.8%,14.3%,and 9.7%in prediction accuracy,recommendation coverage,and diversity,respectively,while also exhibiting advantages in user satisfaction and knowledge mastery metrics.
作者
张泽宇
张颖娇
张宏俊
叶昊
ZHANG Zeyu;ZHANG Yingjiao;ZHANG Hongjun;YE Hao(China Mobile Group Beijing Co.,Ltd.,Beijing 100032;Postal Industry Technology Research and Development Center of State Post Bureau of Nanjing University of Posts and Telecommunications(Internet of Things Technology),Nanjing 210003,China;School of Internet of Things,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;China Telecommunications Services Co.,Ltd.,Beijing 100073,China)
出处
《软件工程》
2025年第10期67-71,共5页
Software Engineering
关键词
个性化学习推荐系统
知识图谱
深度学习
数据稀疏性优化
推荐算法优化
personalized learning recommendation system
knowledge graph
deep learning
data sparsity optimization
recommendation algorithm optimization
