物联网差异数据库中的故障数据快速挖掘仿真.pdfVIP

第 33 卷第 1 期 计算机仿真 2016 年 1 月 文章编号:1伽-9348(2016)01-0301-04 物联网差异数据库中的故障数据快速挖掘仿真 钱晓军，范冬萍，吉根林 {南京师范大学计算机学院，江苏南京 21αYn ) 摘要:对大型物联网环境下的差异数据库中的故障数挺进行准确位测?t.由于大型物联网环槐下差异数据库受到环槐状况 和数据质量的影响，造成故障敖据挖掘时敬锯特征描述不准确。 传统的算法在故障数据挖掘过程中，由于受到不准确符征 的影响，无法准确划分故障费~~与非故降数据的邻域，导致挖掘准确率降低。 提出采用离群点算法的大型物联网环绕下的 差异数据库中的故障数据挖掘方法。 将差异数据库中的数据属性分为故障属性和非故障属性，用故障属性计算放撑邻域， 用非故降属性确定数锯的故降离在羊皮，对差异数据库中的所有的数据的放回事离群程度进行计算，并进行降序排列，从而完成 故障数据的挖掘。 实验结果袋明，利用改进算法进行差异数据库中的故障数据挖掘，能够有效提高挖掘的准确率，降低挖掘 辑时。 关键词:大型物联网;差异数据库;故障数据;挖掘 中图分类号:TP311 文自由标识码 :B DitTerences Between the Internet of Things in the Database Failure Data Rapid Excavation Simulation QIAN Xiao-讪n ， FAN Dong-ping , JI Geng-lin ( College of Computer Nanjing Normal University. Nanjing Jiangsu 2H阴7.China) ABSTRACf: For 阳ge differences under the environment of lntemet of things in the database failure data accurately di回鸣. differences due to the large Intemet environment datahase a.ffected by environmental condilíons and the data quality .ωωe the fault data mining data character d臼criplíon is not a∞urste. The 田ditional a1gorithm in fault in the proc棚。f data mining, due to the effect of the inaccurste features cannot be accurstely classified fault data and fault data of neighborhood. results in the dec陀ase of mining accuracy. Outliers based on the a1gorithm of the large differ- ences in under the environment of Intemet of things in the database failure data mining method. DifIerence attribute data in the database can be divided into fault and the faull attribute . us忧e fau咄1t a创tl时ri二bu a f缸￥u叫ùl prop阱e眈ωie销5 determine the fault d仇egr芭es from lhe 箩0ωu叩p of data. the differences between all of the data in the da- tabase of fault degree is calcu1ated from the group. and in descending order . to complete 由e failu陀 data mining. The e环lerimental results show that the improved a1gorithm difference fault data mining in the database ， ωn effective1y im- prove the accurscy in mining. r琶duce mining time-consuming. KEYWORDS :Large intemet of things; Differences