微信服务号

微信订阅号

2025年5月1日 7:55 星期四
首页 > 过刊浏览>2022年第33卷第10期 >3656-3672. DOI:10.13328/j.cnki.jos.006629
PDF HTML阅读 XML下载 导出引用 引用提醒
突破智慧教育:基于图学习的课程推荐系统
作者:
作者简介:

智慧信息系统新技术专题

通讯作者:

李青山，qshli@mail.xidian.edu.cn

中图分类号:

TP18

基金项目:

国家自然科学基金(61972300,61902288,U21B2015);陕西省自然科学基础研究计划(2020JQ-300);国家重点研发计划(2019YFB1406404);中国科学院战略性先导科技专项(C类)(XDC05040100)


Breakthrough in Smart Education: Course Recommendation System Based on Graph Learning
Author:
  • 摘要
    • | |
  • 访问统计
    • |
  • 参考文献 [54]
    • |
  • 相似文献
    • |
  • 引证文献
    • | |
  • 文章评论
    摘要:

    近年来,随着互联网技术的迅猛发展,以慕课(MOOC)为代表的在线教育平台得到广泛普及.为助力“因材施教”的个性化智慧教育,以推荐算法为代表的人工智能技术受到了学术界与工业界的普遍关注.虽然在电子商务等领域获得了成功应用,但推荐算法与在线教育融合时仍面临严峻挑战:现有算法对隐式交互数据的挖掘不充足,推荐背后的知识指导作用不明显,面向实践的推荐系统软件有缺失.对此,设计了一套面向工业化场景的智慧课程推荐系统:(1)提出基于图卷积神经网络的推荐引擎,将“用户-课程”隐式交互数据建模为异构图;(2)将课程知识信息融入“用户-课程”异构图,深入挖掘了“用户-课程-知识”关联关系;(3)设计了高效的在线推荐系统,实现了“预处理-召回-离线排序-在线推荐-结果融合”的多段流水线原型,不仅能够快速响应课程推荐请求,更能有效缓解推荐算法落地的最大障碍——冷启动问题.最后,基于真实课程学习平台数据集,以对比实验表明了离线推荐引擎相比其他主流推荐算法的先进性,并基于两个典型用例分析验证了在线推荐系统面临工业场景需求的可用性.

    Abstract:

    With the rapid development of Internet technology, online course learning platforms like MOOCs have gained wide popularity in recent years. In order to facilitate the personalized smart education of "teach students in accordance with their aptitude", artificial intelligence technologies, represented by recommendation algorithms, have been widely focused by communities of academia and industry. Although it has been successfully applied in e-commerce and other fields, the integration of recommendation algorithms into online education scenarios still faces severe challenges: Existing algorithms are not competent in mining implicit interactives, guidance from knowledge towards recommendation is not effective, and there are still few sights on practical recommendation system software. Therefore, an intelligent course recommendation system is proposed for industrial scenarios, which includes: (1) an offline recommendation engine based on graph convolutional neural network, modeling the "user-course" implicit interaction behavior as a heterogeneous graph, and extracting course knowledge information to guide the model learning and training. Therefore, the relationship like "user-course-knowledge" can be fully and deeply mined; (2) an efficient online recommendation system prototype based on multi- stage pipeline of "preprocess-recall-offline sort-online recommend-result fuse". Besides quickly responding to course recommendation requests, it can effectively alleviate the major obstacle to recommender systems, the cold start problem. Finally, based on a course learning dataset from real platform, extensive experiments are conducted to show that the proposed offline recommendation engine is competitive with other mainstream recommendation algorithms. And based on the analysis of two typical use cases, the usability of the proposed online recommendation system when facing industrial needs is verified.

    参考文献
    [1] Liu B, Xue NN, Guo HF, Tang RM, Stefanos Z, He XQ, Li ZG. AutoGroup:Automatic feature grouping for modelling explicit high-order feature interactions in CTR prediction. In:Proc. of the 43rd ACM Conf. on Research and Development in Information Retrieval. New York:ACM, 2020. 199-208.
    [2] Shang YM, Cao YN, Liu YB. Combining content-aware and heterogeneous-aware for event recommendation. Ruan Jian Xue Bao/Journal of Software, 2020, 31(4):1212-1224(in Chinese with English abstract). http://www.jos.org.cn/1000-9825/5544.htm[doi:10.13328/j.cnki.jos.005544]
    [3] Xu HL, Wu X, Li XD, Yan BP. Comparison study of Internet recommendation system. Ruan Jian Xue Bao/Journal of Software, 2009, 20(2):350-362(in Chinese with English abstract). http://www.jos.org.cn/1000-9825/3388.htm[doi:10.3724/SP.J.1001.2009. 03388]
    [4] Tang CL, Liao JX, Wang HC, Sung CY, Lin WC. Conceptguide:Supporting online video learning with concept map-based recommendation of learning path. In:Proc. of the Web Conf. Ljubljana:ACM, 2021. 12.
    [5] Wang C, Zhu HS, Zhu C, Zhang X, Chen E, Xiong H. Personalized employee training course recommendation with career development awareness. In:Proc. of the Web Conf. ACM, 2020. 12.
    [6] Asmaa E, George K. Domain-aware grade prediction and top-n course recommendation. In:Proc. of the 10th ACM Conf. on Recommender Systems. New York:ACM, 2016. 183-190.
    [7] Shen M, Lai TP, Wang SM, Peng YM, Gao ZT. Design and implementation of the course recommendation engine in Peking University. CAAI Trans. on Intelligent Systems, 2015, 10(3):369-375(in Chinese with English abstract).
    [8] Liu X, He JR, Sam D, Liz O. Convolution-consistent collective matrix completion. In:Proc. of the 28th ACM Conf. on Information and Knowledge Management. New York:ACM, 2019. 2209-2212.
    [9] Huang LW, Jiang BT, Lv SY, Liu YB, Li DY. Survey on deep learning based recommender system. Chinese Journal of Computers, 2018, 41(7):1619-1647(in Chinese with English abstract).
    [10] Huang YY, Kuo TY, Chen HH. Selecting representative thumbnail image and video clip from a video via bullet screen. In:Proc. of the Web Conf. New York:ACM, 2020. 48-49.
    [11] Xu C, Li Q, Ge JF, Gao JY, Yang XY, Pei CH, Sun F, Wu J, Sun HX, Ou WW. Privileged features distillation at taobao recommendations. In:Proc. of the 26th ACM Conf. on Knowledge Discovery&Data Mining. New York:ACM, 2019. 2590-2598.
    [12] Li XJ, Liu H, Shi HX, Zhu LQ, Zhang YH. Deep learning based course recommendation model. Journal of Zhejiang University (Engineering Science), 2019, 53(11):2139-2145, 2162(in Chinese with English abstract).
    [13] Ge Y, Chen SC. Graph convolutional network for recommender systems. Ruan Jian Xue Bao/Journal of Software, 2020, 31(4):1101-1112(in Chinese with English abstract). http://www.jos.org.cn/1000-9825/5928.htm[doi:10.13328/j.cnki.jos.005928]
    [14] He XN, Deng K, Wang X, Li Y, Zhang YD, Wang M. LightGCN:Simplifying and powering graph convolution network for recommendation. In:Proc. of the 43rd Int'l ACM SIGIR Conf. on Research and Development in Information Retrieval. New York:ACM, 2020. 639-648.
    [15] Paul C, Jay A, Emre S. Deep neural networks for YouTube recommendations. In:Proc. of the 10th ACM Conf. on Recommender Systems. New York:ACM, 2016. 191-198.
    [16] Huang JT, Ashish S, Sun SY, Xia Li, David Z, Philip P, Janani P, Giuseppe O, Yang LJ. Embedding-based retrieval in Facebook search. In:Proc. of the 26th ACM SIGKDD Int'l Conf. on Knowledge Discovery&Data Mining. New York:ACM, 2020. 2553-2561.
    [17] Rex Y, He RN, Chen KF, Pong E, William L, Jure L. Graph convolutional neural networks for Web-scale recommender systems. In:Proc. of the 24th ACM SIGKDD Int'l Conf. on Knowledge Discovery&Data Mining. New York:ACM, 2018. 974-983.
    [18] Liu Q, Xie RB, Chen L, Liu SK, Tu K, Cui P, Zhang B, Lin LY. Graph neural network for tag ranking in tag-enhanced video recommendation. In:Proc. of the 29th ACM Int'l Conf. on Information&Knowledge Management. New York:ACM, 2020. 2613-2620.
    [19] Yehuda K, Robert B, Chris V. Matrix factorization techniques for recommender systems. Computer, 2009, 8(42).
    [20] Liu HF, Jing LP, Yu J. Survey of matrix factorization based recommendation methods by integrating social information. Ruan Jian Xue Bao/Journal of Software, 2018, 29(2):340-362(in Chinese with English abstract). http://www.jos.org.cn/1000-9825/5391.htm[doi:10.13328/j.cnki.jos.005391]
    [21] Yehuda K. Factorization meets the neighborhood:A multifaceted collaborative filtering model. In:Proc. of the 43rd ACM Conf. on Knowledge Discovery&Data Mining. New York:ACM, 2008. 426-434.
    [22] Chen JY, Zhang HW, He XN, Nie LQ, Liu W, Tat-Seng C. Attentive collaborative filtering:Multimedia recommendation with item-and component-level attention. In:Proc. of the 40th ACM Conf. on Research and Development in Information Retrieval. New York:ACM, 2017. 335-344.
    [23] Marco G, Augusto P. ItemRank:A random-walk based scoring algorithm for recommender engines. In:Proc. of the 20th AAAI Conf. on Artificial Intelligence. Hyderabad:AAAI, 2007. 2766-2771.
    [24] Shi C, Hu BB, Zhao WX, Yu PS. Heterogeneous information network embedding for recommendation. IEEE Trans. on Knowledge and Data Engineering, 2019, 31(2):357-370.
    [25] Hamilton WL, Ying R, Leskovec J. Inductive representation learning on large graphs. In:Proc. of the 31st NIPS Conf. on Neural Information Processing Systems. NIPS, 2017. 1025-1035.
    [26] Thomas NK, Max W. Semi-supervised classification with graph convolutional networks. In:Proc. of the 5th ICLR Conf. on Learning Representations. Toulon:ICLR, 2017. 14.
    [27] Wang X, He XN, Wang M, Feng FL, Tat-Seng C. Neural graph collaborative filtering. In:Proc. of the 42nd ACM Conf. on Research and Development in Information Retrieval. New York:ACM, 2019. 165-174.
    [28] Rianne VDB, Thomas NK, Max W. Graph convolutional matrix completion. In:Proc. of the 24th ACM Conf. on Knowledge Discovery&Data Mining. London:ACM, 2018. 9.
    [29] David G, David N, Brian M, Douglas T. Using collaborative filtering to weave an information tapestry. Communications of the ACM, 1992. 61-70.[doi:10.1145/138859.138867]
    [30] Liu K, Li XY, Zhu ZH, Lodewijk B, Wang H. Factor-bounded nonnegative matrix factorization. ACM Trans. on Knowledge Discovery from Data, 2021, 18.
    [31] Lin X, Zhang WP, Zhang M, Zhu WW, Pei J, Zhao PL, Huang JZ. Online compact convexified factorization machine. In:Proc. of the 2018 World Wide Web Conf. ACM, 2018. 1633-1642.
    [32] He XN, Liao LZ, Zhang HW, Nie LQ, Hu X, Chua TS. Neural collaborative filtering. In:Proc. of the 26th Int'l Conf. on World Wide Web. ACM, 2017. 173-182.
    [33] Feng Y, Zhang B, Qiang BH, Zhang YY, Shang JX. MN-HDRM:A novel hybrid dynamic recommendation model based on long-short-term interests multiple neural networks. Chinese Journal of Computers, 2019, 42(1):16-28(in Chinese with English abstract).
    [34] Xu CF, Zhao PP, Liu YC, Xu JJ, Victor S, Cui ZM, Zhou XF, Xiong H. Recurrent convolutional neural network for sequential recommendation. In:Proc. of the World Wide Web Conf. New York:ACM, 2019. 3398-3404.
    [35] Gong JB, Wang S, Wang JL, Feng WZ, Peng H, Tang J, Philip S. Attentional graph convolutional networks for knowledge concept recommendation in MOOCs in a heterogeneous view. In:Proc. of the 43rd Int'l ACM SIGIR Conf. on Research and Development in Information Retrieval. ACM, 2020. 10.
    [36] He XR, Pan JF, Jin O, Xu TB, Liu B, Xu T, Shi YX, Antoine A, Ralf H, Stuart B, Joaquin Q. Practical lessons from predicting clicks on ADS at Facebook. In:Proc. of the 8th Int'l Workshop on Data Mining for Online Advertising. New York:ACM, 2014. 1-9.
    [37] Zhou GR, Zhu XQ, Song CR, Fan Y, Zhu H, Ma X, Yan YH, Jin JQ, Li H, Gai K. Deep interest network for click-through rate prediction. In:Proc. of the 24th ACM SIGKDD Int'l Conf. on Knowledge Discovery&Data Mining. New York:ACM, 2018. 1059-1068.
    [38] Dai J. Bridge the gap:Industrial data pipeline oriented recommendation enhancement. In:Proc. of the 44th Int'l ACM SIGIR Conf. on Research and Development in Information Retrieval. New York:ACM, 2021. 1.
    [39] Fan M, Guo JC, Zhu S, et al. MOBIUS:Towards the next generation of query-ad matching in Baidu's sponsored search. In:Proc. of the 25th ACM SIGKDD Int'l Conf. on Knowledge Discovery. Anchorage:ACM, 2019. 2509-2517.
    [40] Jin BW, Gao C, He XN, Jin DP, Li Y. Multibehavior recommendation with graph convolutional networks. In:Proc. of the 43rd Int'l ACM SIGIR Conf. on Research and Development in Information Retrieval. New York:ACM, 2020. 10.
    [41] Ajit P, Geoffrey J. Relational learning via collective matrix factorization. In:Proc. of the 14th ACM SIGKDD Int'l Conf. on Knowledge Discovery&Data Mining. New York:ACM, 2008. 650-658.
    [42] Zheng Y, Gao C, He X, Li Y, Jin D. Price-aware recommendation with graph convolutional networks. In:Proc. of the 36th IEEE Int'l Conf. on Data Engineering. IEEE, 2020. 133-144.
    [43] Shi Y, Martha L, Alan H. Mining contextual movie similarity with matrix factorization for context-aware recommendation. ACM Trans. on Intelligent Systems and Technology, 2013, 19.
    [44] Wang HW, Zhao M, Xie X, Li WJ, Guo MY. Knowledge graph convolutional networks for recommender systems. In:Proc. of the World Wide Web Conf. New York:ACM, 2019. 3307-3313.
    [45] Liang D, Rahul G, Matthew D, Tony J. Variational autoencoders for collaborative filtering. In:Proc. of the 2018 World Wide Web Conf. ACM, 2019. 689-698.
    附中文参考文献:
    [2] 尚燕敏,曹亚男,刘燕兵.基于异构社交网络信息和内容信息的事件推荐.软件学报, 2020, 31(4):1212-1224. http://www.jos.org.cn/1000-9825/5544.htm[doi:10.13328/j.cnki.jos.005544]
    [3] 许海玲,吴潇,李晓东,阎保平.互联网推荐系统比较研究.软件学报, 2009, 20(2):350-362. http://www.jos.org.cn/1000-9825/3388.htm[doi:10.3724/SP.J.1001.2009.03388]
    [7] 沈苗,来天平,王素美,彭一明,高志同.北京大学课程推荐引擎的设计和实现.智能系统学报, 2015, 10(3):369-375.
    [9] 黄立威,江碧涛,吕守业,刘艳博,李德毅.基于深度学习的推荐系统研究综述.计算机学报, 2018, 41(7):1619-1647.
    [12] 厉小军,柳虹,施寒潇,朱柳青,张亚辉.基于深度学习的课程推荐模型.浙江大学学报(工学版), 2019, 53(11):2139-2145+2162.
    [13] 葛尧,陈松灿.面向推荐系统的图卷积网络.软件学报, 2020, 31(4):1101-1112. http://www.jos.org.cn/1000-9825/5928.htm[doi:10.13328/j.cnki.jos.005928]
    [20] 刘华锋,景丽萍,于剑.融合社交信息的矩阵分解推荐方法研究综述.软件学报, 2018, 29(2):340-362. http://www.jos.org.cn/1000-9825/5391htm[doi:10.13328/j.cnki.jos.005391]
    [33] 冯永,张备,强保华,张逸扬,尚家兴. MN-HDRM:长短兴趣多神经网络混合动态推荐模型.计算机学报, 2019, 42(1):16-28.
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

歹杰,李青山,褚华,周洋涛,杨文勇,卫彪彪.突破智慧教育:基于图学习的课程推荐系统.软件学报,2022,33(10):3656-3672

复制
分享
文章指标
  • 点击次数:2906
  • 下载次数: 5871
  • HTML阅读次数: 3561
  • 引用次数: 0
历史
  • 收稿日期:2021-07-20
  • 最后修改日期:2021-12-24
  • 在线发布日期: 2022-02-22
  • 出版日期: 2022-10-06
文章二维码
友情链接:中国科学院软件研究所中国计算机学会中国科学院国家自然科学基金委员会CCF数字图书馆Int'l J of Softw Info计算机系统应用
您是第19892678位访问者
版权所有：中国科学院软件研究所 京ICP备05046678号-3
地址：北京市海淀区中关村南四街4号,邮政编码：100190
电话：010-62562563 传真：010-62562533 Email：jos@iscas.ac.cn
技术支持：北京勤云科技发展有限公司

京公网安备 11040202500063号