首页 | 本学科首页   官方微博 | 高级检索  
     

基于斜率偏离的时间序列相似性搜索方法研究*
引用本文:梁建海,张建业,杨峰,潘泉. 基于斜率偏离的时间序列相似性搜索方法研究*[J]. 计算机应用研究, 2010, 27(1): 54-55. DOI: 10.3969/j.issn.1001-3695.2010.01.015
作者姓名:梁建海  张建业  杨峰  潘泉
作者单位:1. 西北工业大学,自动化学院,西安,710072
2. 空军工程大学,西安,710051
基金项目:国家自然科学基金资助项目(60634030,60702066);国防预研基金资助项目
摘    要:摘要:针对数据挖掘领域中时间序列的相似性度量问题,提出了基于斜率偏离度量的浮动索引相似性搜索算法。在斜率偏离度量的基础上建立分箱标志,通过浮动索引方法有效实现高维多元时间序列的准确索引,用实际的飞行数据进行仿真验证获得好的效果,证明这种方法的有效性和可靠性。

关 键 词:时间序列; 斜率偏离; 相似性搜索; 浮动索引

Study of distance-measuring method and application for time series similarity search based on slope departure
LIANG Jian-hai,ZHANG Jian-ye,YANG Feng,PAN Quan. Study of distance-measuring method and application for time series similarity search based on slope departure[J]. Application Research of Computers, 2010, 27(1): 54-55. DOI: 10.3969/j.issn.1001-3695.2010.01.015
Authors:LIANG Jian-hai  ZHANG Jian-ye  YANG Feng  PAN Quan
Affiliation:1.School of Automation/a>;Northwestern Polytechnical University/a>;Xi'an 710072/a>;China/a>;2.Air Force Engineering University/a>;Xi'an 710051/a>;China
Abstract:Aiming at the measuring method of similarity search in time series, this paper proposed a similarity measuring method based on slope departure. To dispose the time series data with high dimension, used a similarity measuring method to caculate the slope departure based on curve-in bin solving the localization of slope distance measurement in SFFS. In fact, the physical concept of this method proved more clearly and nicely with real fight data. The maturity, validity and the continuity of slope departure was proved.
Keywords:time series   slope departure   similarity search   SFFS
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《计算机应用研究》浏览原始摘要信息
点击此处可从《计算机应用研究》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号