基于本体的网络学习资源推荐算法设计

赵继春; 孙素芬; 郭建鑫; 钟瑶; 王敏; 秦莹

基于本体的网络学习资源推荐算法设计

赵继春^{1, 2,},
孙素芬^{1, 2},
郭建鑫^{1, 2},
钟瑶^{1, 2},
王敏^{1, 2},
秦莹^{1, 2}

基金项目:

北京市自然科学基金 9172009

北京市科技计划项目 Z191100007519010

北京市农业科技项目 20200127

详细信息

作者简介:
赵继春  男，(1980-)，博士，副研究员.研究方向为机器学习、数据挖掘.E-mail:zhaojichun123@163.com

孙素芬  女，(1964-)，硕士，研究员.研究方向为智能化系统

郭建鑫  男，(1973-)，博士，研究员.研究方向为智能化系统

钟瑶  女，(1993-)，硕士.研究方向为数据挖掘

王敏  女，(1982-)，硕士，助理研究员.研究方向为数据挖掘

秦莹  女，(1982-)，硕士，助理研究员.研究方向为数据挖掘

中图分类号: TP311.5; G431
计量
- 文章访问数: 317
- HTML全文浏览量: 169
- PDF下载量: 20
- 被引次数: 0
出版历程
- 收稿日期: 2020-03-30
- 修回日期: 2020-04-19

Recommendation algorithm design of network learning resource based on ontology

ZHAO Ji-chun^{1, 2
,},
SUN Su-fen^{1, 2},
GUO Jian-xin^{1, 2},
ZHONG Yao^{1, 2},
WANG min^{1, 2},
QIN Ying^{1, 2}

摘要

摘要:
针对学习者在网络学习平台中面临的信息过载导致获取个性化信息不足问题，应用领域本体构建方法，对学习课件进行数据建模及表示，构建了一种基于领域本体与学习者属性信息的特征模型，模型结合领域本体主题及学习者基本信息要素具有语义支持并随时间推移进行更新.以学习特征模型为基础设计一种融合相似度的协同过滤推荐方法，有效缓解推荐算法面临的冷启动及其语义支持不足的问题.实验结果表明，所提的方法与传统的协同过滤推荐方法相比，数据推荐的准确率MAE减小12.29%，有效提高数据推荐质量.
- 本体 /
- 特征模型 /
- 资源推荐 /
- 语义 /
- 个性化 /
- 相似度
Abstract:
In view of the problem that learners are faced with information overload in the network learning platform, which leads to insufficient access to personalized information, this paper applies the domain ontology construction method to model and represent the data of learning courseware, and constructs a learner feature model based on domain ontology and learner attribute information. The model has semantic support combined with domain ontology theme and basic information elements of learners and updates with time. Based on the learning feature model, a collaborative filtering recommendation method is designed to integrate similarity, which can effectively solve the problem of cold start and lack of semantic support. The experimental results show that the data recommendation precision of proposed method is better than the traditional collaborative filtering recommendation method, the accuracy of data recommendation decreased by 12.29%, effectively improving the quality of data recommendation.
- ontology /
- feature model /
- resource recommendation /
- semantics /
- personalization /
- similarity

HTML全文

图 1 学习课件本体主要类关系

下载: 全尺寸图片幻灯片

图 2 学习课件类、类层次和属性逻辑关系

下载: 全尺寸图片幻灯片

图 3 学习者特征模型层次结构

下载: 全尺寸图片幻灯片

图 4 算法设计思路

下载: 全尺寸图片幻灯片

表 1 不同训练集占比条件的SPA、MAE和COV

训练集比例	SPA	方法一		方法二
训练集比例	SPA	MAE	COV	MAE	COV
0.2	0.024 1	0.921 3	0.853 5	0.801 6	0.841 3
0.4	0.036 9	0.905 5	0.872 3	0.784 7	0.871 5
0.6	0.048 3	0.863 3	0.891 3	0.721 3	0.883 4
0.8	0.060 6	0.812 6	0.903 4	0.703 5	0.897 2
平均值		0.875 7	0.880 1	0.752 8	0.873 4

下载: 导出CSV

参考文献(11)

[1]	王光, 姜丽, 董帅含, 等.融合本体语义与用户属性的协同过滤算法[J].计算机工程, 2019, 45(10): 215-220. DOI: 10.19678/j.issn.1000-3428.0052499. WANG G, JIANG L, DONG S H, et al. Collaborative filtering algorithm combining ontology semantics and user attribute[J]. Computer Engineering, 2019, 45(10): 215-220. DOI: 10.19678/j.issn.1000-3428.0052499.
[2]	申云凤.基于多重智能算法的个性化学习路径推荐模型[J].中国电化教育, 2019(11): 67-72. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDJY201911009.htm SHEN Y F. Personalized learning path recommendation model based on multiple intelligent algorithms[J]. China Educational Technology, 2019(11): 67-72. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDJY201911009.htm
[3]	LAWSON C, BEER C, ROSSI D, et al. Identification of 'at Risk' students using learning analytics: the ethical dilemmas of intervention strategies in a higher education institution[J]. Educational Technology Research and Development, 2016, 64(5): 957-968. DOI: 10.1007/s11423-016-9459-0.
[4]	杨桢, 从传锋.基于大数据的网络教育方向选取方法研究[J].现代电子技术, 2018, 41(15): 87-91. DOI: 10.16652/j.issn.1004-373x.2018.15.020. YANG Z, CONG C F. Research on network education direction selection method based on big data[J]. Modern Electronics Technique, 2018, 41(15): 87-91. DOI: 10.16652/j.issn.1004-373x.2018.15.020.
[5]	黄立威, 江碧涛, 吕守业, 等.基于深度学习的推荐系统研究综述[J].计算机学报, 2018, 41(7): 1619-1647. DOI: 10.11897/SP.J.1016.2018.01619. HUANG L W, JIANG B T, LV S Y, et al. Survey on deep learning based recommender systems[J]. Chinese Journal of Computers, 2018, 41(7): 1619-1647. DOI: 10.11897/SP.J.1016.2018.01619.
[6]	何洁月, 马贝.利用社交关系的实值条件受限玻尔兹曼机协同过滤推荐算法[J].计算机学报, 2016(1): 183-195. DOI: 10.11897/SP.J.1016.2016.00183. HE J Y, MA B. Based on real-valued conditional restricted Boltzmann machine and social network for collaborative filtering[J]. Chinese Journal of Computers, 2016(1): 183-195. DOI: 10.11897/SP.J.1016.2016.00183.
[7]	WEI J, HE J H, CHEN K, et al. Collaborative filtering and deep learning based recommendation system for cold start items[J]. Expert Systems with Applications, 2017, 69: 29-39. DOI: 10.1016/j.eswa.2016.09.040.
[8]	ZHAO W X, LI S, HE Y L, et al. Connecting social media to e-commerce: cold-start product recommendation using microblogging information[J]. IEEE Transactions on Knowledge and Data Engineering, 2016, 28(5): 1147-1159. DOI: 10.1109/TKDE.2015.2508816.
[9]	汪方胜, 侯立文, 蒋馥.领域本体建立的方法研究[J].情报科学, 2005, 23(2): 241-244. DOI: 10.3969/j.issn.1007-7634.2005.02.020. WANG F S, HOU L W, JIANG F. Study on establishing domain ontology[J]. Information Science, 2005, 23(2): 241-244. DOI: 10.3969/j.issn.1007-7634.2005.02.020.
[10]	邹锋.基于深度神经网络和改进相似性度量的推荐算法[J].计算机应用与软件, 2019, 36(11): 286-293. https://www.cnki.com.cn/Article/CJFDTOTAL-JYRJ201911047.htm ZOU F. A recommendation algorithm based on deep neural networks and improved similarity measurement[J]. Computer Applications and Software, 2019, 36(11): 286-293. https://www.cnki.com.cn/Article/CJFDTOTAL-JYRJ201911047.htm
[11]	王维, 高岭, 高全力.融合用户信任和用户兴趣漂移的协同过滤算法[J].微电子学与计算机, 2019, 36(7): 103-108. DOI: 10.19304/j.cnki.issn1000-7180.2019.07.020. WANG W, GAO L, GAO Q L. Collaborative filtering algorithm based on user trust and interest drift detecting[J]. Microelectronics & Computer, 2019, 36(7): 103-108. DOI: 10.19304/j.cnki.issn1000-7180.2019.07.020.