基于FastICA和G-G聚类的多元时序自适应分段

王玲; 李泽中

doi:10.12263/DZXB.20220649

您当前的位置：

首页 >

文章列表页 >

基于FastICA和G-G聚类的多元时序自适应分段

学术论文 | 更新时间：2025-12-08

- 基于FastICA和G-G聚类的多元时序自适应分段
- Adaptive Segmentation of Multivariate Time Series with FastICA and G-G Clustering
- 电子学报 2023年51卷第5期页码：1235-1244
- 作者机构：
  
  1.北京科技大学自动化学院，北京100083
  2.北京科技大学自动化学院工业过程知识自动化教育部重点实验室，北京100083
- 作者简介：
  
  [ "王玲　女，1974年出生于北京市.现为北京科技大学自动化学院教授. 研究方向为数据挖掘、模式识别. E-mail: lingwang@ustb.edu.cn" ]
  [ "李泽中　男，1998年出生于山西省.现为北京科技大学自动化学院硕士生. 研究方向为数据挖掘、模式识别.E-mail: g20208697@xs.ustb.edu.cn" ]
- 基金信息：
  
  国家自然科学基金(62076025;61572073)
- DOI：10.12263/DZXB.20220649
  中图分类号： TP273
- 收稿：2022-06-07，
  
  修回：2022-08-22，
  
  纸质出版：2023-05-25
- 稿件说明：
移动端阅览
王玲,李泽中.基于FastICA和G-G聚类的多元时序自适应分段[J].电子学报,2023,51(05):1235-1244.

WANG Ling,LI Ze-zhong.Adaptive Segmentation of Multivariate Time Series with FastICA and G-G Clustering[J].ACTA ELECTRONICA SINICA,2023,51(05):1235-1244.
王玲,李泽中.基于FastICA和G-G聚类的多元时序自适应分段[J].电子学报,2023,51(05):1235-1244. DOI： 10.12263/DZXB.20220649.

WANG Ling,LI Ze-zhong.Adaptive Segmentation of Multivariate Time Series with FastICA and G-G Clustering[J].ACTA ELECTRONICA SINICA,2023,51(05):1235-1244. DOI： 10.12263/DZXB.20220649.

摘要

现有多元时间序列的分段方法主要通过检测时序数据统计特性或形状的变化情况，并以此为依据对分段点的位置进行“硬划分”.然而，这些分段方法无法对两个分段之间的过渡区间长度进行准确估计，且普遍需要人为预先设置参数，在高维且噪声较强的情况下分段效果较差.本文针对现有分段方法存在的诸多不足，提出一种基于FastICA（Fast Independent Component Analysis）和G-G（Gath-Geva）模糊聚类的多元时序自适应分段方法.该方法利用FastICA进行特征提取，采用DW（Durbin-Watson）指数自动选取高信噪比的主成分，并根据最小描述长度（Minimum Description Length，MDL）设计基于G-G模糊聚类的自适应分段模型，实现对于多元时间序列的“软划分”.基于多种领域的真实数据集实验结果表明：与现有主流的分段方法相比，本文方法在上述数据集上的平均

和MAE（Mean Absolute Error）可分别提升8.4%~16.8%和3.06%~6.56%.

Abstract

The existing segmentation methods detect the statistical or shape changes of multivariate time series

and perform crisp segmentation on the location of change points. However

these methods fail to estimate the length of the transition interval between two segments

cannot accurately segment multivariate time series with high dimension

strong noise

and need to set parameters in advance. To address such matters

an adaptive multivariate time series segmentation method based on FastICA (Fast Independent Component Analysis) and G-G (Gath-Geva) clustering is proposed. In this method

the key features of multivariate time series are extracted via FastICA

and DW (Durbin-Watson) criterion is used to automatically select main components with high signal-to-noise ratio. According to the minimum description length (MDL)

an adaptive multivariate time series segmentation model based on G-G clustering is designed

which is able to perform soft segmentation of multivariate time series. The experimental analysis is carried out on real datasets in many different fields. Compared with state-of-art benchmarks

the average

and MAE (Mean Absolute Error) of the proposed method on the above-mentioned datasets improve 8.4%~16.8% and 3.06%~6.56%

respectively.

关键词

Keywords

references

MAYA S , YAMAGUCHI A , NISHINO K , et al . Lag-aware multivariate time-series segmentation [C]// Proceedings of the 2020 SIAM International Conference on Data Mining . Philadelphia : SIAM , 2020 : 622 - 630 .

XU J , ZHOU J , TAN P N , et al . Spatio-temporal multi-task learning via tensor decomposition [J]. IEEE Transactions on Knowledge and Data Engineering , 2019 , 33 ( 6 ): 2764 - 2775 .

CHE Z , PURUSHOTHAM S , CHO K , et al . Recurrent neural networks for multivariate time series with missing values [J]. Scientific Reports , 2018 , 8 ( 1 ): 1 - 12 .

KEOGH E , CHU S , HRT D , et al . An online algorithm for segmenting time series [C]// Proceedings 2001 IEEE International Conference on Data Mining . Piscataway : IEEE , 2001 : 289 - 296 .

CHUNG F L , FU T C , NG V , et al . An evolutionary approach to pattern-based time series segmentation [J]. IEEE Transactions on Evolutionary Computation , 2004 , 8 ( 5 ): 471 - 489 .

LIU S , YAMADA M , COLLIER N , et al . Change-point detection in time-series data by relative density-ratio estimation [J]. Neural Networks , 2013 , 43 : 72 - 83 .

WANG P , WANG H , WANG W . Finding semantics in time series [C]// Proceedings of the 2011 ACM SIGMOD International Conference on Management of Data . New York : ACM , 2011 : 385 - 396 .

LI L , MCCANN J , POLLARD N S , et al . Dynammo: mining and summarization of coevolving sequences with missing values [C]// Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining . New York : ACM , 2009 : 507 - 516 .

HALLAC D , NYSTRUP P , BOYD S . Greedy gaussian segmentation of multivariate time series [J]. Advances in Data Analysis and Classification , 2018 , 13 ( 3 ): 727 - 751 .

MATSUBARA Y , SAKURAI Y , FALOUTSOS C . Autoplait: automatic mining of co-evolving time sequences [C]// Proceedings of the 2014 ACM SIGMOD International Conference on Management of Data . New York : ACM , 2014 : 193 - 204 .

LIM H , CHOI H , CHOI Y , et al . Memetic algorithm for multivariate time-series segmentation [J]. Pattern Recognition Letters , 2020 , ( 138 ): 60 - 67 .

KAWABATA Y , MATSUBARA Y , SAKURAI Y . Automatic sequential pattern mining in data streams [C]// Proceedings of the 28th ACM International Conference on Information and Knowledge Management . New York : ACM , 2019 : 1733 - 1742 .

GHARGHABI S , YEH C C M , Ding Y , et al . Domain agnostic online semantic segmentation for multi-dimensional time series [J]. Data Mining and Knowledge Discovery , 2019 , 33 ( 1 ): 96 - 130 .

GHARGHABI S , DING Y , YEH C C M , et al . Matrix profile VIII: domain agnostic online semantic segmentation at superhuman performance levels [C]// 2017 IEEE International Conference on Data Mining . Piscataway : IEEE , 2017 : 117 - 126 .

DELDARI S , SMITH D V , SADRI A , et al . Espresso: Entropy and shape aware time-series segmentation for processing heterogeneous sensor data [J]. Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies , 2020 , 4 ( 3 ): 1 - 24 .

HALLAC D , VARE S , BOYD S , et al . Toeplitz inverse covariance-based clustering of multivariate time series data [C]// Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining . New York : ACM , 2017 : 215 - 223 .

王玲 , 朱慧 . 基于KPCA和G-G聚类的多元时间序列模糊分段 [J]. 控制与决策 , 2021 , 36 ( 1 ): 115 - 124 .

WANG Ling , ZHU Hui . Fuzzy segmentation of multivariate time series with KPCA and G-G clustering [J]. Control and Decision , 2021 , 36 ( 1 ): 115 - 124 . (in Chinese)

ZHOU F , DE LA TORRE F , HODGINS J K . Hierarchical aligned cluster analysis for temporal clustering of human motion [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2013 , 35 ( 3 ): 582 - 596 .

HOAI M , DE LA TORRE F . Maximum margin temporal clustering [C]// Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics (AISTAT) . Cambridge : JMLR , 2012 : 520 - 528 .

GRUNWALD P D . The Minimum Description Length Principle [M]. Cambridge : MIT Press , 2007 : 132 .

HYVARINEN A , OJA E . Independent component analysis: algorithms and applications [J]. Neural Networks , 2000 , 13 ( 4 ): 411 - 430 .

HYVARINEN A . Fast and robust fixed-point algorithms for independent component analysis [J]. IEEE Transactions on Neural Networks , 1999 , 3 ( 10 ): 626 - 634 .

ABONYI J , FEIL B , NEMETH S , et al . Modified gath–geva clustering for fuzzy segmentation of multivariate time-series [J]. Fuzzy Sets and Systems , 2005 , 149 ( 1 ): 39 - 56 .

BOUVERESSE D J R , RUTLEDGE D N . Independent Components Analysis: Theory and Applications [M]// Data Handling in Science and Technology Resolving Spectral Mixtures with Applications from Ultrafast Time-Resolved Spectroscopy to Super-Resolution Imaging . 1st ed . New York : Elsevier , 2016 : 225 - 277 .

FENG W , LIU S , FALOUTSOS C , et al . Eaglemine: vision-guided micro-clusters recognition and collective anomaly detection [J]. Future Generation Computer Systems , 2021 , ( 115 ): 236 - 250 .

LIN J F S , KARG M , KULIC D . Movement primitive segmentation for human motion modeling: A framework for analysis [J]. IEEE Transactions on Human-Machine Systems , 2016 , 46 ( 3 ): 325 - 339 .

HODGINS J K . CMU graphics lab motion capture database [DB/OL]. [ 2020-06-27 ]. http://mocap.cs.cmu.edu/ http://mocap.cs.cmu.edu/ .

BARBIC J , SAFONOVA A , PAN J Y , et al . Segmenting motion capture data into distinct behaviors [C]// Proceedings of Graphics Interface 2004 . Waterloo : Canadian Human-Computer Communications Society , 2004 : 185 - 194 .

REINHARDT A , BAUMANN P , BURGSTAHLER D , et al . On the accuracy of appliance identification based on distributed load metering data [C]// 2012 Sustainable Internet and ICT for Sustainability (SustainIT) . Piscataway : IEEE , 2012 : 1 - 9 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据