社交网络下学习推荐研究与实践

Research on Learning Recommendation System and Its Implementation on Academic Social Networks

学习推荐系统为学习者快速获取学习所需材料和资源提供支持,社交网络兴起使在线用户的互动和行为更丰富,对推动用户学习效率提高有积极作用。以建构主义学习理论为基础,考虑学习者基本属性、学习风格、所在学习环境等因素,利用本体论技术,结合过滤推荐算法,提出了社交网络环境下基于本体的学习推荐系统架构和功能;构建了学习者本体,加强其特征描述,并管理学习过程中产生的学习者本体流;以本体语义相似度计算方法为基础,提出相似学习者查找机制,实现学习资源推荐;依托学术社交网络Scholat,开发实现了基于本体的学习推荐系统,试运行显示该系统能够较有效推荐学习资源,引导学习进展,有利于学习者学习效率的提高。

Learning Recommended System supports the fast access to learning materials and resources for learners, while social networks can rich online learners＇ interaction and behaviors, so that to promote learning efficiency. Considering the learners＇ basic properties, learning pattern, environment and other factors, we used the ontology technology, combined with filtering algorithm based on the constmctivism learning theory; eventually we proposed a learning recommendation system architecture and functionality on the academic social network. We constructed learners＇ ontology to strengthen its characterization, and manage the learners＇ ontology stream which generated during learning process; we proposed a mechanism to find similar learners based on the ontology semantic similarity. Using the scholar social website Scholat, we designed the ontology-based learning recommended system and implemented to get the effective combination of interaction. Finally, an experiment of course learning is put into practice, the result shows that the system can recommend more effective learning resources, guide the learning progress, and improve learning efficiency.