A Review and Evaluation of Elastic Distance Functions for Time Series Clustering
A Review and Evaluation of Elastic Distance Functions for Time Series Clustering
复制标题
DOI：
10.1007/s10115-023-01952-0
复制
发表时间：
2022-05
期刊：
影响因子：
--
通讯作者：
Christopher Holder;Matthew Middlehurst;A. Bagnall
中科院分区：
文献类型：
--
作者：
Christopher Holder;Matthew Middlehurst;A. Bagnall
时间序列聚类是对时间序列数据进行分组而不依赖于标签的行为。聚类时间序列的算法可以分为两组：采用时间序列特定距离度量的算法和从时间序列中获取特征的算法。这两种方法通常都依赖于传统的聚类算法，例如ask-means。我们的重点是分区聚类算法，采用弹性距离措施，即距离，同时测量距离进行某种重新排列。我们描述了九种常用的弹性距离度量，并将它们的性能与k-means和k-medoids聚类器进行了比较。我们的发现，基于使用UCR时间序列档案的实验，是令人惊讶的。我们发现，一般来说，DTW距离的聚类并不比使用欧几里得距离更好，采用编辑结合翘曲的距离测量明显优于其他方法。我们进一步观察到，使用k-medoids聚类器，而不是感谢手段，提高了所有九个弹性距离措施的聚类。一个函数，移动-分割-合并（MSM）距离，是本研究中性能最好的算法，时间扭曲编辑（TWE）距离紧随其后。我们的结论是，MSM或TWE与k-medoids聚类器应被认为是一个很好的替代DTW的聚类时间序列的弹性距离措施。我们提供了实现，广泛的结果和在相关GitHub存储库上复制结果的指导。
Time series clustering is the act of grouping time series data without recourse to a label. Algorithms that cluster time series can be classified into two groups: those that employ a time series specific distance measure and those that derive features from time series. Both approaches usually rely on traditional clustering algorithms such ask-means. Our focus is on partitional clustering algorithms that employ elastic distance measures, i.e. distances that perform some kind of realignment whilst measuring distance. We describe nine commonly used elastic distance measures and compare their performance withk-means andk-medoids clusterer. Our findings, based on experiments using the UCR time series archive, are surprising. We find that, generally, clustering with DTW distance is not better than using Euclidean distance and that distance measures that employ editing in conjunction with warping are significantly better than other approaches. We further observe that usingk-medoids clusterer rather thank-means improves the clusterings for all nine elastic distance measures. One function, the move–split–merge (MSM) distance, is the best performing algorithm of this study, with time warp edit (TWE) distance a close second. Our conclusion is that MSM or TWE withk-medoids clusterer should be considered as a good alternative to DTW for clustering time series with elastic distance measures. We provide implementations, extensive results and guidance on reproducing results on the associated GitHub repository.