基于机器学习算法的校园网学生上网行为评估方法研究  

作　　者：李骞[1] 王硕[1] 隋继学[1] 

机构地区：[1]河南牧业经济学院能源与智能工程学院,河南郑州450011

出　　处：《科技创新与应用》2021年第33期1-5,11,共6页Technology Innovation and Application

基　　金：2020年河南牧业经济学院博士科研启动资金项目(编号:2020HNUAHEDF008)。

摘　　要：随着互联网的迅速发展,大学生上网时间和频率呈指数上升趋势。在大数据环境背景下,作为教育工作者和学生管理者,如何通过上网数据了解大学生真实的学习生活情况,培养学生良好的网络习惯是高等教育质量提升的新机遇和挑战。文章提出了一种在线获取学生网络流量及上网日志的模型,利用随机森林(Random Forest,RF)和梯度增强决策树(Gradient Boosting Decision Tree,GBDT)两种机器学习方法进行数据分析,实现对学生上网行为的准确评估。并且为指导学生养成良好的网络习惯提出相应的建议和对策,以求提高高等教育的质量。With the rapid development of the Internet,the online time and frequency of college students are increasing exponentially.In the context of big data environment,how to understand the real study and life of college students through online data and cultivate students'good network habits for an educator and counselor has posed challenge and brings a new opportunity to the improvement of the quality of higher education.In this paper,a model is proposed to obtain students'network traffic and online logs.Random forest(RF)and gradient boosting decision tree(GBDT)are used for data analysis,so as to accurately evaluate students'online behavior.Finally,through extracting and analyzing the log data of students'online behavior,the corresponding suggestions and countermeasures are put forward,in order to improve the quality of higher education.

关 键 词：机器学习 网络数据分析 网络爬虫 梯度提升决策树 随机森林 学生上网行为 

分 类 号：S767.5[农业科学—森林保护学]