基于改进模糊均值聚类的汉语言学习用户学习行为模式研究  

Study on the Learning Behavior Model of Chinese Language Learners Based on Improved Fuzzy C-means Clustering

作　　者：年晴 NIAN Qing(School of Humanities Management,Shanxi University of Traditional Chinese Medicine,Xi’an 712046,China)

机构地区：[1]陕西中医药大学人文管理学院,陕西西安712046

出　　处：《微型电脑应用》2020年第10期69-72,共4页Microcomputer Applications

基　　金：2019年陕西中医药大学校级教育教学改革研究项目(2019jg19)。

摘　　要：为了实现汉语言在线学习用户学习行为聚类分析,针对FCM聚类结果易受其初始聚类中心选择的影响,提出一种基于IHS-FCM的汉语言学习用户学习行为聚类分析。选择参与维度、专注维度、规律维度、交互维度和学习成绩等作为学习行为的分析指标,学习者层次分为5个等级,分别为优秀、良好、中、合格和差。与HS-FCM、SVM和决策树对比发现,文中算法IHS-FCM具有更高的聚类准确率和更快的收敛速度以及更低的适应度,为学习者层次划分和优化课程学习提供了新的方法。In order to analyze the learning behavior of Chinese language online learners,due to FCM clustering results are easily affected by the initial cluster center selection,this paper presents a clustering analysis of Chinese language learning behavior based on IHS-FCM.Participation dimension,attention dimension,regularity dimension,interaction dimension and learning achievement are selected as the analysis indexes of learning behavior.The learners'level is divided into 5 grades,i.e.,excellent,good,medium,qualified and poor.Compared with HS-FCM,SVM and decision tree,the IHS-FCM has higher clustering accuracy,faster convergence speed and lower fitness.It provides a new method for learners to divide and optimize course learning.

关 键 词：在线学习 汉语言 学习行为 模糊均值聚类 和声搜索算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]