[1]王玲艳 李婵 张梦琪.基于DeepWalk的在线教育平台智能学习资源推荐研究[J].信息化理论与实践,2022,02(02):204-224.
　Research on Intelligent Learning Resource Recommendation for Online Education Platform Based on DeepWalk Zhang Mengqi 1,2 Wang Lingyan1,2 Li Chan 3[J].Information Theory and Practice,2022,02(02):204-224.
点击复制

基于DeepWalk的在线教育平台智能学习资源推荐研究()
分享到：

《信息化理论与实践》[ISSN:2520-5862/CN:]

卷:
02
期数:
2022年02
页码:
204-224
栏目:
出版日期:
2023-12-30

文章信息/Info

Title:
Research on Intelligent Learning Resource Recommendation for Online Education Platform Based on DeepWalk Zhang Mengqi 1,2 Wang Lingyan1,2 Li Chan 3
作者:
王玲艳1; 2 李婵3 张梦琪
1(福州大学信息管理研究所 福建福州 350108 )
Author(s):
1
(Institute of Information Management, Fuzhou University, Fuzhou Fujian 350108, China) 2 ( Fuzhou University Library, Fuzhou Fujian 350108 , China) 3 (School of Public Management, Fujian Agriculture and Forestry University, Fuzhou Fujian 350108, China)
关键词:
DeepWalk SkipGram语言模型 学习资源推荐
Keywords:
DeepWalk SkipGram language model Learning resource recommendation
摘要:
目的]本文将蕴含特征表示能力的深度随机游走方法引入在线学习推荐领域，挖掘学习者偏好特征从而进行智能学习资源推荐。[方法] 该方法利用学习者评分数据构造的二部图生成随机游走顶点序列，并利用SkipGram语言模型获取图顶点的低维嵌入向量表示。然后结合协同过滤的思想，使用学习者向量表示计算学习者间的相似度，从而进行在线课程个性化推荐。[结果]利用中国大学MOOC在线教育平台真实数据集将本文方法与基线方法进行对比，实验表明本文提出方法效果更优。[局限]本文仅挖掘学习者对课程的用户偏好，未来可考虑结合信任机制，引入信任网络，建模用户偏好模型，获得更加准确的学习者特征表示。[结论]深度游走模型可以分析学习者的行为偏好，有效提升在线学习资源个性化推荐效果。
Abstract:
Objective] In this paper, the DeepWalk method with feature representation ability is introduced into the field of online learning recommendation, so as to mine learners’ personal preferences and make intelligent recommendations. [Methods] According to the interactive data between learners and online courses, a bipartite graph model was constructed. Random walks are carried out on bipartite graphs to generate vertex sequences, and the neighbor structure model of graph vertices is captured by using SkipGram language model to learn the low-dimensional embedding vector representation of graph vertices. Calculate the similarity between learners’ vertices, and use collaborative filtering idea to make personalized recommendation of online courses. [Results] The real data set of MOOC online education platform of China University was used to compare the proposed method with the baseline method, and the experiment showed that the proposed method was more effective. [Limitations] This paper only explores learners’ preferences for courses. In the future, we can consider introducing trust network to model users’ preferences in combination with trust mechanism to obtain a more accurate representation of learners’ characteristics. [Conclusions] The DeepWalk model can analyze learners’ behavior preferences and effectively improve the personalized recommendation effect of online learning resources.

参考文献/References:

[1]董永峰,王雅琮,董瑶,等.在线学习资源推荐综述[J/OL].计算机应用:1-10[2023-02-25].

http://kns.cnki.net/kcms/detail/51.1307.tp.20221107.1453.004.html

[2]刘芳,田枫,李欣,等.融入学习者模型在线学习资源协同过滤推荐方法[J].智能系统学报,2021,16(06):1117-1125.

[3]赵继春,孙素芬,郭建鑫,等.农业在线学习资源知识图谱构建与推荐技术研究[J].计算机应用与软件,2022,39(08):69-75.

[4]Tahir S, Hafeez Y , Abbas M A , et al. Smart Learning Objects Retrieval for E-Learning with Contextual Recommendation based on Collaborative Filtering[J]. Education and Information Technologies, 2022: 1-38.

[5]Mbipom B , Craw S , Massie S. Improving e‐learning recommendation by using background knowledge[J]. Expert Systems,2021,38(7): e12265.

[6]Ibrahim T S, Saleh A I , Elgaml N , et al. A fog based recommendation system for promoting the performance of E-Learning environments[J]. Computers & Electrical Engineering,2020,87: 106791.

[7] Perozzi B, Al- Rfou R, Skiena S. Deepwalk: Online learning of social representations[C]//Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining. 2014: 701-710.

[8]崔冬,王明,李刚,等.基于多级深度特征与随机游走的显著性检测[J].华南理工大学学报(自然科学版),2020,48(08):49-55.

[9]Yang?K,?Zhu?J.?Next?poi?recommendation?via?graph?embedding?representation?from?h-deepwalk?on?hybrid?network[J].?IEEE?Access,?2019,?7:?171105-171113.

[10]Qu Z, Li B, Wang X, et al. An efficient recommendation framework on social media platforms based on deep learning[C]//2018 IEEE International Conference on Big Data and Smart Computing (BigComp). IEEE, 2018: 599-602.

[11]Cai L, Wang J, He T, et al. A novel link prediction algorithm based on deepwalk and clustering method[C]//Journal of Physics: Conference Series. IOP Publishing, 2018, 1069(1): 012131.

[12]刘靖凯.基于深度随机游走的协同过滤推荐算法[J].科学技术创新,2021(06):93-94.

[13]冯曦,朱福喜,刘世超.基于深度游走模型的标签传播社区发现算法[J].计算机工程,2018,44(03):220-225 .

更新日期/Last Update: 2024-06-20