基于大规模在线开放课程的学习者模型的设计与实现  

Design and realization of student model based on massive open online courses

作　　者：黄丹霞 刘欣欣[1] HUANG Danxia;LIU Xinxin(School of Computer Science&Engineering,South China University of Technology,Guangzhou Guangdong 510006,China)

机构地区：[1]华南理工大学计算机科学与工程学院,广州510006

出　　处：《计算机应用》2018年第A02期327-330,共4页journal of Computer Applications

基　　金：广东省高等教育教学改革项目(GDJG20142052);华南理工大学教研教改项目(Y1180781)

摘　　要：针对大规模在线开放课程(MOOC)学生对学习过程参与不足、缺乏个性化的学习指导等问题,提出利用智能教学系统的学习者模型的方法,构建MOOC学习者模型。首先针对MOOC平台收集的学习行为数据的特点,选取学习者特征;然后,基于贝叶斯网络构建知识跟踪模型,基于经验概率设置模型参数,并在模型中引入问题的难度;最后,定义学生态度积极性特征,基于分类算法构建学习态度跟踪模型。在MOOC数据集上对模型进行实验,对比了不同贝叶斯知识跟踪模型预测的准确率,当采用逻辑回归算法作为学生态度积极性分类算法时,可较准确地预测学习态度。实验结果表明,该学习者模型具备一定的对学生的知识水平和态度进行分析的能力。Concerning the problems in Massive Open Online Course(MOOC)that students are not involved enough in the learning process and no personalized learning guidance is provided,the learner model of intelligent tutoring system was used to build an MOOC learner model.Firstly,learners’characteristics were collected according to the learning behavior data collected by MOOC platform;secondly,a knowledge tracing model based on Bayesian network was constructed.Model parameters were set based on their empirical probabilities and the difficulty of a problem was introduced into the model.Finally,the characteristics of learner’s attitude and enthusiasm were defined,and a learning attitude tracing model was designed based on classification algorithms.In the experiments conducted on a dataset of MOOC,the predictions of different Bayesian knowledge tracing models were compared.When Logistic regression algorithm was used as the classifier of students’attitude and enthusiasm,the learning attitude could be predicted accurately.The experimental results show that the MOOC learner model is capable of analyzing students’knowledge level and attitude.

关 键 词：大规模在线开放课程 智能教学系统 学习者模型 贝叶斯网络 分类算法 逻辑回归 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]