在线学习结果预测与激励——基于学习者行为建模  

The Prediction and Motivation of Online Learning Results--Based on Learner Behavior Modeling

作　　者：丁晟春[1] 周轩屹 王玉东[1] 俞沣洋 DING Shengchun;ZHOU Xuanyi;WANG Yudong;YU Fengyang

机构地区：[1]南京理工大学经济管理学院

出　　处：《吉林省教育学院学报》2023年第12期55-59,共5页Journal of Jilin Provincial Institute of Education

基　　金：江苏省高校教学研究工作专项课题“基于学习者行为建模的混合式教学新模式理论与实践应用”(2021JSJY074);南京理工大学高等教育教学改革研究课题“Python语言系列课程建设与改革”(2021-A-18)。

摘　　要：在线课程普遍具有注册率高、通过率低的问题。为了提高在线课程通过率,本文从在线学习者参与课程的测试、作业、讨论角度,分析在线学习课程通过者、未通过者的学习行为,对比两类学习者的学习行为特征,筛选两类学习者的特征,并根据所选特征建立机器学习模型,对比分析SVM模型和BP神经网络模型,对在线课程学习结果进行预测;最后,基于加涅九阶段教学设计,提出具体干预措施,并结合预测模型对提出措施进行实施效果验证,结果表明干预措施对提高在线课程通过率具有一定的效果。Online courses generally have the problems of high registration rate and low pass rate.In order to improve the pass rate of online courses,this paper analyzes the learning behaviors of those who pass online learning courses and those who fail online learning courses,compares the characteristics of learning behaviors of the two types of learners,screens the characteristics of the two types of learners,and builds a machine learning model based on the selected characteristics.Predictions of online learning results are made by comparing and analyzing SVM model and BP neural network models.Finally,based on the nine stage teaching design of Gagne,specific intervention measures are proposed,and the implementation effect of the proposed measures is verified by combining the prediction model.The results show that the intervention measures have a certain effect on improving thepass rateofonlinecourses.

关 键 词：学习者行为 在线学习结果预测 激励措施 神经网络 SVM 

分 类 号：G434[文化科学—教育学]