计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (5): 35-44.doi: 10.11896/jsjkx.230200074

• 数据库&大数据&数据科学 • 上一篇    下一篇

基于对抗策略类别特定的多样性时间序列shapelets提取

罗颖, 万源, 王礼勤   

  1. 武汉理工大学理学院 武汉 430070
  • 收稿日期:2023-02-13 修回日期:2023-07-13 出版日期:2024-05-15 发布日期:2024-05-08
  • 通讯作者: 万源(wanyuan@whut.edu.cn)
  • 作者简介:(319552@whut.edu.cn)
  • 基金资助:
    中央高校基本科研业务费专项资金(2021III030JC)

Category-specific and Diverse Shapelets Extraction for Time Series Based on Adversarial Strategies

LUO Ying, WAN Yuan, WANG Liqin   

  1. School of Science,Wuhan University of Technology,Wuhan 430070,China
  • Received:2023-02-13 Revised:2023-07-13 Online:2024-05-15 Published:2024-05-08
  • About author:LUO Ying,born in 1999,postgraduate.Her main research interests include data mining,machine learning and feature selection.
    WAN Yuan,born in 1976,Ph.D,professor.Her main research interests include data mining,pattern recognition,manifold learning,machine learning and feature selection.
  • Supported by:
    Fundamental Research Funds for the Central Universities of Ministry of Education of China(2021III030JC).

PDF (PC)

362

摘要： 在时间序列分类任务中,通过提取时间序列的shapelets进行分类的方法因分类准确率高且具有良好的可解释性而受到广泛关注。针对现有方法学习到的shapelets是所有类共享,可以区分大多数类但不能准确地区分某一类和其他类,以及使用对抗策略的模型生成的shapelets存在多样性不足等问题,提出了一种基于对抗策略类别特定的多样性时间序列shapelets提取方法。该方法将类别信息嵌入时间序列,采用多生成器模块对抗地生成多个有差别的类别特定shapelets,再通过施加差异约束来提高shapelets的多样性,最后使用shapelet转换得到的特征对时间序列进行分类。在36个时间序列数据集上与5种基于shapelets的算法和11种先进的分类算法进行实验对比,实验结果表明,所提方法分别在36个数据集中的26个和20个数据集上取得了最优结果,且均取得了最高的平均秩,平均分类准确率相比其他方法最少提高了2.4%,最多提高了17.8%。消融性分析以及可视化分析验证了多样性和类别特定的思路在时间序列分类上的有效性。

关键词: 时间序列, shapelets, 类别特定, 多样性, 对抗网络

Abstract: For time series classification,the method of classification by extracting the shapelets of time series and has attracted widespread attention due to its high classification accuracy and good interpretability.Most of the existing shapelets-based me-thods learn the shared shapelets for all classes,which can distinguish most classes,but not the unique class.Besides,the shapelets obtained by those models using adversarial strategies have problems like insufficient diversity.In order to solve these problems,this paper proposes a category-specific and diverse shapelets extraction method based on adversarial strategies.This method embeds the category information into the time series,adversarially generates a number of different category-specific shapelets by using the multi-generator module.The diversity of shapelets are guaranteed by imposing a difference constraint,and the last step uses the features obtained by the shapelets transformation to classify the time series.The proposed method is experimentally compared with 5 shapelets-based algorithms and 11 state-of-the-art classification algorithms on 36 time-series datasets.Experimental results show that,compared with 5 shapelets-based algorithms and 11 advanced classification algorithms,the proposed method achieves the best results on 26 and 20 datasets out of 36 datasets,and both achieve the highest average ranks,and its ave-rage classification accuracy is 2.4% higher than other methods at least,and 20% higher at most.Ablation analysis and visualization analysis demonstrate the effectiveness of diversity and category-specific approaches to time series classification.

Key words: Time series, shapelets, Category-specific, Diversity, Adversarial networks

中图分类号: 

  • TP391
[1]ARUL M,KAREEMA.Applications of Shapelet Transform toTime Series Classification of Earthquake,Wind and Wave Data[J].Engineering Structures,2021,228:111564.
[2]AHMED T,SINGH D.Probability Density Functions BasedClassification of MODIS NDVI Time Series Data and Monitoring of Vegetation Growth Cycle[J].Advances in Space Research,2020,66(4):873-886.
[3]Al-HADEETHI H,ABDULLA S,DIYKH M,et al.Adaptive Boost LS-SVM Classification Approach for Time-Series Signal Classification in Epileptic Seizure Diagnosis Applications[J].Expert Systems with Applications,2020,161:113676.
[4]YEUNG J F K A,WEI Z,CHAN K Y,et al.Jump Detection in Financial Time Series Using Machine Learning Algorithms[J].Soft Computing,2020,24(3):1789-1801.
[5]RATANAMAHATANA C A,KEOGH E J.Three Myths aboutDynamic Time Warping Data Mining[C]//Proceedings of the 2005 SIAM International Conference on Data Mining.California:SIAM,2005:506-510.
[6]DING H,TRAJCEVSKI G,SCHEUERMANN P,et al.Que-rying and Mining of Time Series Data:Experimental Comparisonof Representations and Distance Measures[J].Proceedings of the VLDB Endowment,2008,1(2):1542-1552.
[7]YE L,KEOGH E.Time Series Shapelets:A New Primitive for Data Mining[C]//Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.2009:947-956.
[8]MUEEN A,KEOGH E,YOUNG N.Logical-Shapelets:An Ex-pressive Primitive for Time Series Classification[C]//Procee-dings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York:Association for Computing Machinery,2011:1154-1162.
[9]RAKTHANMANON T,KEOGH E.Fast Shapelets:A Scalable Algorithm for Discovering Time Series Shapelets[C]//Procee-dings of the 2013 SIAM International Conference on Data Mi-ning.2013:668-676.
[10]ZHANG Z G,ZHANG H W,WEN Y L,et al.Discriminative extraction of features from time series[J].Neurocomputing,2018,275:2317-2328.
[11]LINES J,DAVIS L M,HILLS J,et al.A shapelet transform for time series classification[C]//Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mi-ning.Austin:SIAM,2012:289-297.
[12]GRABOCKA J,SCHILLING N,WISTUBA M,et al.Learning Time-Series Shapelets[C]//Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York:Association for Computing Machinery,2014:392-401.
[13]HOU L,KWOK J,ZURADA J.Efficient Learning of TimeseriesShapelets[C]//Proceedings of the AAAI Conference on Artificial Intelligence.Phoenix:AAAI Press,2016:1209-1215.
[14]LI G,CHOI B,XU J,et al.Efficient Shapelet Discovery for Time Series Classification[J].IEEE Transactions on Knowledge and Data Engineering,2020,34(3):1149-1163.
[15]WANG Z,YAN W,OATES T.Time Series Classification from Scratch with Deep Neural Networks:A Strong Baseline[C]//2017 International Joint Conference on Neural Networks.Anchorage:IEEE,2017:1578-1585.
[16]WANG Y,EMONET R,FROMONT E,et al.Learning Interpretable Shapelets for Time Series Classification through Adversarial Regularization [J].arXiv:1906.00917,2019.
[17]MA Q,ZHUANG W,LI S,et al.Adversarial Dynamic Shapelet Networks[C]//Proceedings of the AAAI Conference on Artificial Intelligence.New York:AAAI Press,2020:5069-5076.
[18]BOUBRAHIMI S F,HAMDI S M,MA R,et al.On the Mining of the Minimal Set of Time Series Data Shapelets[C]//2020 IEEE International Conference on Big Data(Big Data).Atlanta:IEEE,2020:493-502.
[19]LIN J,KEOGH E,WEI L,et al.Experiencing SAX:A Novel Symbolic Representation of Time Series[J].Data Mining and Knowledge Discovery,2007,15(2):107-144.
[20]GOODFELLOW I,POUGET-ABADIE J,MIRZA M,et al.Ge-nerative adversarial nets[C]//Advances in Neural Information Processing Systems 27,Montreal,Quebec,Canada.2014:2672-2680.
[21]MIRZA M,OSINDERO S.Conditional Generative AdversarialNets[J].arXiv:1411.1784,2014.
[22]ODENA A.Semi-Supervised Learning with Generative Adversarial Networks[C]//Proceedings of the 33nd International Conference on Machine Learning.New York:JMLR,2016.
[23]ODENA A,OLAH C,SHLENS J.Conditional Image Synthesis with Auxiliary Classifier GANs[C]//Proceedings of the 34th International Conference on Machine Learning.Sydney:JMLR,2017:2642-2651.
[24]GHOSH A,KULHARIA V,NAMBOODIRI V,et al.Multi-Agent Diverse Generative Adversarial Networks[C]//Confe-rence on Computer Vision and Pattern Recognition.Salt Lake City:IEEE,2018:8513-8521.
[25]KINGMA D P,BA J.Adam:A Method for Stochastic Optimization[J].arXiv:1412.6980,2014.
[26]GÓRECKI T,ŁUCZAK M.Using derivatives in time series classification[J].Data Mining and Knowledge Discovery,2013,26:310-331.
[27]GÓRECKI T,ŁUCZAK M.Non-isometric transforms in timeseries classification using dtw[J].Knowledge-Based Systems,2014,61:98-108.
[28]MIDDLEHURST M,VICKERS W,BAGNALL A.Scalable Dictionary Classifiers for Time Series Classification[C]//Procee-dings of Intelligent Data Engineering and Automated Learning,Lecture Notes in Computer Science.Manchester:Springer,2019:11-19.
[29]SCHÄFER P,LESER U.Fast and Accurate Time Series Classification with WEASEL[C]//Proceedings of the ACM on Conference on Information and Knowledge Management.Singapore:ACM,2017:637-646.
[30]LUBBA C,SETHI S,KNAUTE P,et al.catch22:CAnonicalTime-series CHaracteristics[J].Data Mining and Knowledge Discovery,2019,33(6):1821-1852.
[31]LINES J,TAYLOR S,BAGNALL A.HIVE-COTE:The Hie-rarchical Vote Collective of Transformation Based Ensembles for Time Series Classification [C]//2016 IEEE 16th International Conference on Data Mining(ICDM).Barcelona:IEEE,2016:1041-1046.
[32]LINES J,TAYLOR S,BAGNALL A.Time Series Classificationwith HIVE-COTE:The Hierarchical Vote Collective of Transformation-Based Ensembles [J].ACM Transactions on Know-ledge Discovery from Data,2018,12(5):1-35.
[33]FAWAZ H I,LUCAS B,FPRESTIER G,et al.Inception-time:Finding AlexNet for time series classification[J].Data Mining and Knowledge Discovery,2020,34:1936-1962.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发