计算机集成制造系统 ›› 2016, Vol. 22 ›› Issue (第12期): 2930-2936.DOI: 10.13196/j.cims.2016.12.021

• 产品创新开发技术 • 上一篇    

基于制造云服务QoS序列特性的缺失值估计算法

李珊,俞瑛,胡康华,宋波,姚叶慧   

  1. 南京航空航天大学经济与管理学院
  • 出版日期:2016-12-31 发布日期:2016-12-31
  • 基金资助:
    江苏省自然科学基金资助项目(BK2012385);博士点基金资助项目(20123218120034);南京航空航天大学基本科研业务费资助项目(NS2013083)。

Missing value estimating algorithm based on cloud manufacturing services QoS time series data properties

  • Online:2016-12-31 Published:2016-12-31
  • Supported by:
    Project supported by the Jiangsu Provincial Natural Science Foundation,China(No.BK2012385),the Doctoral Program of Higher Education,China (No.20123218120034),and the Fundamental Research Funds for the Central Universities,China(No.NS2013083).

摘要: 针对目前采用的传统时间序列缺失值估计算法对制造云服务QoS中时间序列缺失数据填补效果不佳的现状,提出了一种新算法。该算法通过考虑制造云服务QoS序列的指标关联性与候选服务间的指标数据相似性,构建了基于服务内部QoS指标关联性的缺失值估计算法和基于候选服务之间指标数据相似性的缺失值估计算法,并将这两种算法通过折衷系数进行融合,得到QoS时间序列缺失值的最优估计值。与传统的时间序列缺失值估计算法进行分析比较,实验结果表明了所提算法的有效性。

关键词: 制造云服务, 服务质量, 时间序列, 缺失值

Abstract: Owing to the problems that traditional missing value estimating algorithms of cloud manufacturing services Quality of Service (QoS) time series did not fill the missing data of time series well,a new algorithm was presented to handle the challenge.By considering two characteristics of QoS time series which called QoS data correlation and QoS data similarity,the missing values estimation algorithm based on QoS data correlation and the missing values estimating algorithm based on QoS data similarity were constructed,and the QoS optimal estimating of missing values was obtained.Compared with traditional missing value estimating algorithms,experimental results showed that the algorithm was an effective algorithm.

Key words: cloud manufacturing services, quality of service, time series, missing value

中图分类号: