电信科学 ›› 2015, Vol. 31 ›› Issue (8): 63-71.doi: 10.11959/j.issn.1000-0801.2015196

• 研究与开发 • 上一篇    下一篇

一种基于LSH的时间子序列匹配查询算法

刘根平,陈叶芳,杜呈透,钱江波   

  1. 宁波大学信息科学与工程学院 宁波 315211
  • 出版日期:2015-08-27 发布日期:2015-08-27
  • 基金资助:
    国家自然科学基金资助项目;浙江省自然科学基金资助项目;宁波市自然科学基金资助项目;宁波市自然科学基金资助项目;“信息与通信工程”浙江省重中之重学科开放基金资助项目

An LSH Based Time Subsequence Matching Algorithm

Genping Liu,Yefang Chen,Chengtou Du,Jiangbo Qian   

  1. Faculty of Electrical and Computer Science,Ningbo University,Ningbo 315211,China
  • Online:2015-08-27 Published:2015-08-27
  • Supported by:
    The National Natural Science Foundation of China;The Natural Science Foundation of Zhejiang Province;The Natural Science Foundation of Ningbo City;The Natural Science Foundation of Ningbo City;Zhejiang Key Discipline Fund of Information and Communication Engineering

摘要:

提出了一种基于LSH(locality sensitive hashing,局部敏感散列)算法处理时间子序列匹配问题的方法LSHSM。不同于FRM和DualMatch方法,该方法不需要对时间序列做DFT、DWT等特征变换,而是直接把序列看成高维数据点,利用LSH能处理高维数据的特性来查找相似时间子序列。实验采用3种不同的时间序列数据集,通过与线性扫描算法比较,验证了算法的有效性,性能有很大的提高。

关键词: 时间子序列, LSH, 匹配查询

Abstract:

An algorithm called LSHSM,which uses locality sensitive hashing(LSH)to process time subsequence matching,was proposed.Different to the FRM and DualMatch algorithms,the LSHSM does not require feature transformation such as DFT and DWT.It just directly regards the sequence as a high-dimensional object to find similar subsequences.Comparing to a linear algorithm on three real datasets,the LSHSM algorithm demonstrates the effectiveness and efficiency.

Key words: time subsequence, locality sensitive hashing, match searching

No Suggested Reading articles found!