摘要
如何有效挖掘学生行为数据是提升学生信息化管理水平的重要内容。针对目前学生信息化管理平台不完善、挖掘精度低的问题,结合决策树、神经网络以及朴素贝叶斯算法建立组合模型,建立基于Spark的学生行为分析与预测平台;同时,以学生消费规律、生活习惯以及学习情况等校园行为作为大数据来源,进行预测分析和实例验证。结果表明:该模型预测结果与实际情况相吻合,平均预测误差不超过5%,验证了所用方法的有效性,可根据学生行为特性分析其行为规律,指导学生行为向全面健康方向发展。
How to effectively mine students’ behavior data is an important content to improve students’ information management level.Aiming at the problems of imperfect student information management platform and low mining accuracy,a combination model based on decision tree,neural network and naive Bayesian algorithm is established to analyze and predict student behavior based on Spark.At the same time,campus behavior such as students’ consumption law,living habits and learning situation is taken as an example.Large data sources are used for forecasting analysis and case validation.The results show that the prediction results of the model are consistent with the actual situation,and the average prediction error is no more than 5% which verifies the model.The law of students’ behavior can be analyzed according to the characteristics of students’ behavior,and the students’ behavior can be guided to develop in an all-round and healthy way.
作者
李铁波
LI Tiebo(Jilin Communications Polytechnic,Changchun 130000,China)
出处
《重庆理工大学学报（自然科学）》
CAS
北大核心
2019年第7期201-206,共6页
Journal of Chongqing University of Technology：Natural Science
基金
国家自然科学基金资助项目(31672348)
关键词
数据挖掘
校园大数据
学生行为
预测模型
决策树
data mining
campus big data
student behavior
prediction model
decision tree
