基于动态时间弯曲的区间值时间序列匹配算法

Similarity matching algorithm for interval-valued time series based on dynamic time warping

结合vague集的优点,将区间值的思想应用于时间序列研究,定义了基于区间值的时间序列.同时,在综合分析了不同的vague值(区间值)的相似度量的缺点后,提出了一种新的基于vague值的相似度量,并结合动态时间弯曲的思想,给出了一个基于动态时间弯曲的区间值时间序列的相似性匹配算法.

This paper studies the time series from the perspective of vague set. That is interval value, rather than point value, is used for representing time series. The main contributions of this paper are as follows. Firstly, the definition of interval-valued time series is introduced. Secondly, by analyzing related work for measuring similarity of interval values in detail, a new similarity measure based on vague set is proposed. Thirdly, a new similarity matching algorithm for interval -valued time series based on dynamic time warping （ DTW ） is proposed.