时间序列数据挖掘相似性度量综述.PDFVIP

第32 卷 第1 期 控 制 与 决 策 Vol. 32 No. 1 2017 年 1 月 Control and Decision Jan. 2017 文章编号: 1001-0920 (2017) 01-0001-11 DOI: 10.13195/j.kzyjc.2016.0462 时间序列数据挖掘的相似性度量综述 a,b a a 陈海燕 , 刘晨晖, 孙 博 (南京航空航天大学a. 计算机科学与技术学院，b. 软件新技术与产业化协同创新中心，南京210016) 摘 要: 在时间序列数据挖掘中, 时间序列相似性是一个重要的概念. 对于诸多算法而言, 能否与一种合适的相似性 度量方法结合应用, 对其挖掘性能有着关键影响. 然而, 至今仍没有统一的度量相似性的方法. 对此, 首先综述了常用 的相似性度量方法, 分析了各自的优点与不足; 其次, 讨论了近年来出现的时序相似性的新解释及其度量方法; 再次, 探讨了相似性度量在时序挖掘任务中的应用以及与挖掘精度的关系; 最后给出了关于时序相似性度量进一步的研究 方向. 关键词: 时间序列数据挖掘；时间序列相似性；相似性度量；挖掘精度 中图分类号: TP273 文献标志码: A Survey on similarity measurement of time series data mining a,b a a CHEN Hai-yan , LIU Chen-hui , SUN Bo (a. School of Computer Science and Technology，b. Collaborative Innovation Center of Novel Software Technology and Industrialization ，Nanjing University of Aeronautics and Astronautics ，Nanjing 210016 ，China) Abstract: Similarity measure is an important concept in time series data mining. For many data mining algorithms, whether it can be used in combination with a suitable time series similarity measure method has a key influence on mining performance. However, there is no uniform definition and measure of similarity. Therefore, we first introduce the most popular similarity measures, and analyze the advantages and disadvantages of each measure. Then, the new interpretations of the time series similarity and the corresponding measures are discussed. Furthermore, we analyze the applications of similarity measures in clustering, classification and regression of time series data, and the relationship between