数据与计算发展前沿 ›› 2022, Vol. 4 ›› Issue (3): 3-18.

CSTR: 32002.14.jfdc.CN10-1649/TP.2022.03.001

doi: 10.11871/jfdc.issn.2096-742X.2022.03.001

• 专刊:先进智能计算平台及应用(下) • 上一篇    下一篇

基于知识图谱技术的线上教学资源推荐系统设计与实现

罗婕溪(),刘帅(),张玉志(),李正丹(),孙羽菲(),张圣林()   

  1. 南开大学,软件学院,天津 300350
  • 收稿日期:2022-02-06 出版日期:2022-06-20 发布日期:2022-06-20
  • 通讯作者: 张玉志
  • 作者简介:罗婕溪,南开大学,硕士研究生,主要研究方向为推荐系统、知识图谱、软件工程等。
    本文主要承担知识图谱推荐系统的总体设计与实现,图嵌入推荐部分的写作。
    LUO Jiexi is a graduate student of Nankai University. Her main research interests include recommendation system, knowledge graph, software engineering, etc.
    In this paper, she is mainly responsible for the overall design and implementation of the knowledge graph recommendation system and the writing of graph embedding recommendation method.
    E-mail: thevolga@163.com|刘帅,南开大学,硕士研究生,研究方向为自然语言处理、软件工程、知识图谱可视化等。
    本文中主要承担知识图谱构建、学习路径推荐和前后置知识点推荐部分的撰写。
    LIU Shuai is a graduate student of Nankai University. His research interests include natural language processing, software engineering and knowledge graph visualization.
    In this paper, he is mainly responsible for the parts of the knowledge graph construction, the recommendation strategy of associative learning path and the recommendation strategy of knowledge in series.
    E-mail: 978951827@qq.com|张玉志,南开大学,讲席教授,博士,软件学院院长,主要研究方向为人工智能、模式识别、自然语言处理等。
    南开大学教学资源平台项目负责人。
    ZHANG Yuzhi, Ph.D., is the chair pro-fessor and the Dean of software college at Nankai University. His research interests focus on artificial intelligence, pattern recognition, natural language processing, etc.
    He is the project leader of the Education resource platform of Nankai University.
    E-mail: zyz@nankai.edu.cn|李正丹,南开大学,助理实验师,硕士,研究方向为人工智能、软件工程等。
    南开大学教学资源平台应用开发负责人,本文主要承担论文修改与审核相关工作。
    LI Zhengdan is an assistant experiment-alist at Nankai University. Her research interests include artif-icial intelligence, software engineering, etc.
    She is the development leader of the Education resource plat-form of Nankai University. She is mainly responsible for the revision and review of the paper.
    E-mail: lzd@nankai.edu.cn|孙羽菲,南开大学软件学院特聘研究员,博士,主要研究方向为深度学习、异构计算、人工智能等。
    本文主要承担论文修改与审核相关工作。
    SUN Yufei, Ph.D., is a professor at Coll-ege of Software, Nankai University. Her research interests include deep learning, heterogeneous computing, artificial intel-ligence, etc.
    She is mainly responsible for the revision and review of the paper.
    E-mail: yufei_sun@sina.com|张圣林,南开大学,软件学院,副教授,博士,主要研究方向为基于机器学习的智能运维,包括异常检测、故障定位、根因分析和故障预测等。
    本文主要承担论文修改与审核相关工作。
    ZHANG Shenglin, Ph.D., is an associate professor at the College of Software, Nankai University. His research interests focus on AIOps, including anomaly detection, failure diagnosis, root cause analysis, failure prediction, etc.
    He is mainly responsible for the revision and review of the paper.
    E-mail: zhangsl@nankai.edu.cn

Online Educational Resources Recommendation System Based on Knowledge Graph Technology

LUO Jiexi(),LIU Shuai(),ZHANG Yuzhi(),LI Zhengdan(),SUN Yufei(),ZHANG Shenglin()   

  1. Department of Software, Nankai University, Tianjin 300350, China
  • Received:2022-02-06 Online:2022-06-20 Published:2022-06-20
  • Contact: ZHANG Yuzhi

RichHTML

51

PDF

247

摘要：

【目的】本文立足于线上教学资源领域,利用知识图谱相关技术,设计并实现一种融合多种推荐策略的推荐系统。【应用背景】线上教学资源种类繁杂、数量众多,且缺乏规范化构建和系统化管理,为教师管理教学资源及学习者查找有效信息造成诸多不便。【方法】通过设计知识图谱数据结构,构建全学科内知识语义关联,融合图嵌入和规则抽取的方法,实现教学资源推荐。【结果】本文提出的线上教学资源推荐系统已应用于南开大学教学资源网平台,在多学科领域内的知识点推荐中展现出较好的效果。【结论】通过实验结果分析,验证了本文推荐系统在缓解“长尾效应”方面的有效性,并通过可视化案例分析,验证了其实用性。

关键词: 知识图谱, 教学资源, 推荐系统, 图嵌入

Abstract:

[Objective] Based on the field of online educational resources, this paper designs and implements a mixed recommendation system of educational resources by knowledge graph technology. [Context] Online educational resources are diverse and numerous and lack standardized construction and systematic management. It is inconvenient for the teachers to manage educational resources and for the learners to find effective information. [Methods] By designing knowledge graph data structure, we construct knowledge semantic association of the whole discipline and integrate graph embedding and rule extraction to realize the educational resource recommendation. [Results] The proposed methodology in this paper has been applied to the educational resource online platform of Nankai University, which achieves remarkable results in the recommendation of knowledge points in multi-disciplinary fields. [Conclusions] By analyzing the experimental results, it turns out that the proposed recommendation system can alleviate the "long tail effect" to some extent, and its practicability is verified by visual case analysis.

Key words: knowledge graph, educational resource, recommendation system, graph embedding