基于分段极值DTW距离的时间序列相似性度量

Similarity Measure in Time Series Based on Segmented Extreme Value Dynamic Time Warping Distance

在时间序列相似性的研究中,通常采用的欧氏距离及其变形无法对在时间轴上发生伸缩或弯曲的序列进行相似性度量,本文提出了一种基于分段极值DTW距离的时间序列相似性度量方法可以解决这一问题。在动态时间弯曲(DTW)距离的基础上,本文定义了序列的分段极值DTW距离,并阐述了其完整的算法实现。与传统的DTW距离相比,分段极值DTW距离在保证度量准确性的同时大大提高了相似性计算的效率。文中最后运用MATLAB作对比实验,并给出实验结果数据,验证了该度量方法的有效性与准确性。

Similarity Measure in time series databases is an important task. Most research work on comparing time series are based on Euclidean distance or its transformations. However Euclidean distance measure will not be an effective method to the time series by scaling and warping along the time-axis. Dynamic time warping （DTW） distance is a good way to deal with these cases, but its large computing limits its application. In this paper, a new method of similarity measure based on segmented extreme value dynamic time warping （SEDTW） distance is put forward. It divides time series into several segments and extracts the extreme values in each segment, and then measuring the new extreme value series on the dynamic time warping distance. Compared with the classical dynamic time distance, this new method is much more fast in speed and almost no degrade in accuracy. This conclusion can also be proved by the experiments in this paper.