基于加权贝叶斯网络的新能源交易数据深度挖掘算法  

Deep mining algorithm of new energy transaction data based on weighted Bayesian network

作　　者：刘涛 LIU Tao(CNO0C New Energy Branch,Beijing 102299)

机构地区：[1]中海石油(中国)有限公司北京新能源分公司,北京102299

出　　处：《长江信息通信》2023年第12期35-37,41,共4页Changjiang Information & Communications

摘　　要：由于新能源交易数据特征不同、数量较大,导致新能源交易数据深度挖掘效果较差,由此,设计一个基于加权贝叶斯网络的新能源交易数据深度挖掘算法。提取数据特征,基于贝叶斯网络,设计TAN分类器。计算其中最大特征值和对应的特征向量,对各个数据加权处理,得到条件特征值。在数据融合后,完成数据深度挖掘,实验结果表明,该算法有效地提高了挖掘准确率与挖掘效率,满足设计需求。Due to the different characteristics and large quantity of new energy transaction data,the deep mining effect of new energy transaction data is poor.Therefore,a weighted Bayesian network based deep mining algorithm for new energy transaction data is designed.Extract data features and design a TAN classifier based on Bayesian networks.Calculate the maximum eigenvalues and corresponding eigenvectors,weight each data,and obtain conditional eigenvalues.After data fusion,complete deep data mining,The experimental results show that the algorithm effectively improves mining accuracy and efficiency,meeting design requirements.

关 键 词：加权贝叶斯网络 新能源交易数据 深度挖掘 特征数据 分类 节点更新 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]