摘要
以山东理工大学数学与统计学院统计学专业2015级学生的7门主干学科成绩作为数据样本,通过贝叶斯网络的结构学习直观地得到了7门学科之间的关联性;基于贝叶斯网络拓扑结构进行了网络的参数学习,得到了各学科成绩的条件概率;最后利用贝叶斯网络推理的联合树算法给出了学生的成绩预测,并用实例证明了贝叶斯网络成绩预测的可行性。
In this paper,we take scores of seven major courses from class 2015 students majoring in statistics in School of Mathematics and Statistics at Shandong University of Technology as the data sample.We intuitively obtain the correlation between various disciplines through the structural learning of the Bayesian network.Then,we also conduct Bayesian network parameter learning based on the Bayesian network topology and obtained conditional probabilities between achievements of various disciplines.Finally,we used the joint tree algorithm of Bayesian network reasoning to predict students′performance.The feasibility of applying Bayesian network to predict student achievement is proved by an example.
作者
刘艳杰
李霞
LIU Yanjie;LI Xia(School of Mathematics and Statistics,Shandong University of Technology,Zibo 255049,China)
出处
《山东理工大学学报（自然科学版）》
CAS
2019年第5期75-78,共4页
Journal of Shandong University of Technology:Natural Science Edition
关键词
贝叶斯网络
K2算法
联合树算法
Bayesian network
K2 algorithm
joint tree algorithm
作者简介
第一作者:刘艳杰,女,1471205869@qq.com;通信作者:李霞,女,18765330861@163.com
