《计算机应用研究》|Application Research of Computers

基于行为序列分析的学习资源推荐算法研究

Research on learning resource recommendation algorithm based on behavior sequence analysis

免费全文下载 (已被下载 次)  
获取PDF全文
作者 桂忠艳,张艳明,李巍巍
机构 黑龙江中医药大学 医学信息工程学院,哈尔滨 150040
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2020)07-012-1979-04
DOI 10.19734/j.issn.1001-3695.2018.12.0930
摘要 利用数据挖掘技术分析网络学习行为数据可以挖掘出其隐含的行为规律特征,为学习者提供个性化的学习资源服务。针对现有的数据挖掘算法在对网络学习行为数据进行分析时普遍存在模型适用性不高的问题,提出了一种基于行为序列分析的学习资源推荐算法。首先,提出行为序列及其相关概念的定义,并提出行为序列相似度计算方法;然后提出基于行为序列相似度的协同过滤推荐算法,计算学习者相似度并为待推荐学习者生成学习资源推荐列表;接着给出基于学习风格的推荐方法,将学习者学习风格特征融入推荐过程;最后,给出基于行为序列分析的学习资源推荐算法的模型。提出的算法没有对行为序列的模式进行限制,具有较高的适用性,对深入研究网络学习行为序列数据为学习者提供个性化学习服务具有一定的借鉴作用。
关键词 网络学习行为; 行为序列相似度; 学习者相似度; 协同过滤; 学习风格
基金项目 黑龙江中医药大学教育教学研究项目(XJJY2018028)
黑龙江省高等教育教学改革研究一般研究项目(SJGY20180486)
本文URL http://www.arocmag.com/article/01-2020-07-012.html
英文标题 Research on learning resource recommendation algorithm based on behavior sequence analysis
作者英文名 Gui Zhongyan, Zhang Yanming, Li Weiwei
机构英文名 College of Medical Information Engineering,Heilongjiang University of Chinese Medicine,Harbin 150040,China
英文摘要 Using the data mining technology to analyze the network learning behavior data can dig out its hidden behavior characteristics, and provide personalized learning resource services for learners. Aiming at the problem that the model applicability of the existing data mining algorithms generally was not very high when analyzing of network learning behavior data, this paper proposed a learning resource recommendation algorithm based on behavior sequence analysis. Firstly, this paper proposed the definition of behavior sequence and its related concepts. Secondly, it proposed the calculation method of behavior sequence similarity. Then, it proposed the collaborative filtering recommendation algorithm based on behavioral sequence similarity to calculate the similarity between learners and to generate the learning resource recommendation list for the learner to be recommended. And then it gave the recommendation algorithm based on learning style, and integrated learners' learning style characteristics into the recommendation process. Finally, it presented the learning resource recommendation algorithm based on behavior sequence analysis. The proposed algorithm does not limit the pattern of behavior sequence, and has a high applicability, it can be used for further study of the network learning behavior data and provide personalized learning service for learners.
英文关键词 network learning behavior; behavior sequences similarity; similarity between learners; collaborative filtering; learning style
参考文献 查看稿件参考文献
 
收稿日期 2018/12/19
修回日期 2019/2/15
页码 1979-1982
中图分类号 TP181
文献标志码 A