基于行为分析的在线课程成绩预测模型  

Online Course Grade Prediction Model Based on Behavior Analysis

作　　者：任占广 尚福华[2] REN Zhan-guang;SHANG Fu-hua(School of Software Engineering,Chongqing University of Arts and Sciences,Chongqing 402160,China;School of Computer&Information Technology,Northeast Petroleum University,Daqing 163318,China)

机构地区：[1]重庆文理学院软件工程学院,重庆402160 [2]东北石油大学计算机与信息技术学院,黑龙江大庆163318

出　　处：《计算机技术与发展》2019年第11期139-143,共5页Computer Technology and Development

基　　金：国家自然科学基金(61170132);国家科技重大专项资助项目(2017ZX05019005-006);重庆文理学院校级科研项目(Z2015RJ05)

摘　　要：随着大学在线课程占全部课题比重的不断提高,为了更加科学地分析在线学习行为和准确地预测在线课程成绩,提出了一种基于行为分析的在线课程成绩预测模型。首先,对学习行为及成绩预测策略进行了系统分析,构建了在线平台数据处理、成绩预测算法设计、成绩预测及算法优化的在线成绩预测机制;其次,利用数据挖掘技术收集在线学习行为数据,结合在线用户的操作特点对行为数据进行分析,提取了与成绩密切相关的10种行为指标数据并存储到数据库中;最后,以“玩课网”平台的重庆文理学院“大学生计算机基础”课程后台数据库作为实验数据基础,结合该课程实施特点,分析了学生学习行为,确定了学习行为指标等级,提取和转换了学生学习行为数据,并利用神经网络实现了在线课程成绩的预测。实验结果表明成绩预测的准确率较高。With the increasing proportion of online courses in all subjects,in order to analyze online learning behaviors more scientifically and predict online course grade more accurately,a behavior analysis-based online course grade prediction model is proposed.Firstly,the learning behavior and grade prediction strategy are systematically analyzed,and an online grade prediction mechanism is constructed,which includes data processing of online platform,design of grade prediction algorithm and optimization of grade prediction algorithm.Secondly,data mining technology is used to collect online learning behavior data,and according to the online user’s operational characteristics,the behavior data is analyzed,and 10 kinds of behavior indicators are extracted and stored in the database.Finally,with the database of the course“Computer Basis for College”of Chongqing University of Arts and Sciences as the platform of“Wankewang”,combined with the characteristics of course implementation,we analyze the learning behavior of students,determine the level of learning behavior indicators,extract and transform the data of students’learning behavior,and utilize the neural network to finish the prediction of online course grade.The experiment shows that the performance prediction is more accurate.

关 键 词：行为分析 数据处理 神经网络 成绩预测 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]