摘要
新冠疫情迫使高等教育的线下教学转向在线学习,教师和学生都不得不适应在线学习平台。因此,对学生学习绩效进行评价对于保证学习质量是至关重要的。本文从学习行为分析、问卷调查和学习状态分析等3个层次对学习绩效进行全面分析。同时使用多分类支持向量机和跨领域迁移学习方法等多种机器学习方法实现对互联网大数据的识别和处理。近2年的教学实践表明,通过对学生在线学习模式下学习绩效的评价可以提高学生对课程的满意度,尽可能减少疫情对学生课程学习的影响。
The COVID-19 pandemic has forced offline teaching in higher education to shift to online learning,and both teachers and students have had to adapt to online learning platforms.Therefore,it is very important to evaluate students’learning performance to ensure the quality of learning.This study conducts a comprehensive analysis of learning performance from three levels of learning behavior analysis,questionnaire survey and student learning status analysis.Learning behavior data is processed using multi classification support vector machine.The research object of learning state analysis is the collected facial expression images of students,and the cross domain transfer learning method is used to realize data identification and processing.The teaching practice in the past two years shows that the evaluation of students’learning performance in the online learning mode can improve students’satisfaction with courses and minimize the impact of the epidemic on students’course learning.
作者
倪彤光
顾晓清
薛磊
NI Tongguang;GU Xiaoqing;XUE Lei(School of Computer Science and Artificial Intelligence,Changzhou University,Changzhou,China,213164)
出处
《福建电脑》
2023年第1期22-26,共5页
Journal of Fujian Computer
基金
江苏省教育科学“十三五”规划课题(No.B-a/2018/01/41)
江苏省教育厅未来网络科研基金项目(No.FNSRFP-2021-YB-36)
常州大学教育教学研究课题(No.GJY2014033)资助。
关键词
在线学习
学习绩效
学习行为分析
机器学习
Online Learning
Learning Performance
Learning Behavior Analysis
Machine Learning
作者简介
通信作者:倪彤光,男,1978年生,研究方向为机器学习。E-mail:ntg@cczu.edu.cn;顾晓清,女,1981年生,研究方向为模式识别;薛磊,女,1969年生,研究方向为教育理论。
