检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:罗杨洋 韩锡斌[1] LUO Yangyang;HAN Xibin(Institute of Education,Tsinghua University,Beijing 100084)
出 处:《电化教育研究》2021年第7期83-90,共8页E-education Research
基 金:国家社会科学基金“十三五”规划2018年度国家一般课题“混合教学的理论体系建构及实证研究”(课题编号:BCA180084)。
摘 要:基于在线学习行为预测学生成绩可以辅助教师动态掌握学情,制定差异化的教学策略,然而在混合课程中仅仅依据在线数据对学生成绩进行预测难度很大,尚处于探索中。文章选取某高校2018秋季学期和2020春季学期的“高活跃型混合课程”学生在线行为数据,采用增量学习的随机森林算法构建学生成绩预测模型,研究发现:(1)增量学习随机森林算法在混合课程样本最多的数据集中,获得预测结果准确率最高(75.1%);(2)相较于批量学习随机森林算法,增量学习算法在数据样本量较多的数据集中预测结果准确率更高;(3)当样本数量达到一定规模后,预测结果准确率波动减小、稳定性增强。本研究采用增量学习随机森林算法预测混合课程中的学生成绩,不仅取得了较好的预测准确率,而且解决了新增数据后模型的稳定性问题,将有助于模型的迭代优化,提高模型的通用性,以及可持续追踪学生在不同学期的学习行为特征。Predicting academic performance based on students'online learning behavior can help teachers dynamically grasp learning conditions and develop differentiated teaching strategies.However,predicting students'academic performance in hybrid courses based on online data is difficult and is still under exploration.In this paper,the online behavioral data of students in the"highly active hybrid courses"in the fall semester 2018 and spring semester 2020 of a university are selected,and the random forest algorithm based on incremental learning is adopted to construct a student performance prediction model.The research findings are as follows:(1)the incremental learning random forest algorithm achieves the highest prediction accuracy(75.1%)in the dataset with the largest sample of hybrid courses.(2)Compared with batch learning random forest algorithm,the incremental learning algorithm has higher accuracy in predicting results in data sets with a large number of data samples.(3)When the number of samples reaches a certain size,the accuracy of the prediction results becomes less volatile and more stable.This study uses the incremental learning random forest algorithm to predict the academic performance of students in hybrid courses,which not only achieves a good prediction accuracy,but also solves the stability problem of the model after adding the new data.It will contribute to the iterative optimization of the model,improve the generality of the model,and continuously track the characteristics of students'learning behaviors in different semesters.
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:216.73.216.116