摘要
由于分区域中异常数据较为分散,现有方法无法对分区域数据集进行有效分类,导致异常数据挖掘效果不理想。为此提出基于朴素贝叶斯的分区域异常数据挖掘方法。获取朴素贝叶斯优化公式,采用假象空间重构数据向量间的欧几里获取度量策略,计算各数据偏差比结果。对分区域中异常数据进行分类,获取相应集合式与触发式。根据数据节点概率化瞬态计算实现分区域异常数据的有效挖掘。仿真结果表明,研究方法分区域异常数据挖掘效率较高,且应用精度更理想,具有较好的实际应用价值。
Due to the scattered abnormal data in subregion, current methods cannot effectively classify the data sets in subregion, resulting in unsatisfactory result. Therefore, a method of mining sub-regional anomaly data based on Naive Bayes was proposed. The optimization formula of Naive Bayes was obtained. The pseudomorph space was used to reconstruct Euclidean between data vectors, and thus to obtain the measurement strategy. Moreover, the deviation ratio of data was calculated. The abnormal data in subregion was classified to obtain the corresponding aggregation and trigger. Based on the probabilistic transient calculation of data node, the effective mining for sub-regional abnormal data was realized. Simulation results prove that the proposed method is efficient and accurate in mining sub-regional anomaly data. This method has good practical value.
作者
康耀龙
冯丽露
张景安
KANG Yao-long;FENG Li-lu;ZHANG Jing-an(School of Computer and Network Engineering,Shanxi Datong University,Datong Shanxi 037009,China;School of Educational Science and Technology,Shanxi Datong University,Datong Shanxi 037009,China;Network Information Center,Shanxi Datong University,Datong Shanxi 037009,China)
出处
《计算机仿真》
北大核心
2020年第10期303-306,316,共5页
Computer Simulation
基金
大同市经济和信息化委员会专项基金项目(JXW2017001)
山西大同大学教学改革创新项目(XJG2019202)
山西省教育科学“十三五”规划课题(GH-18044)。
