ISM: intra-class similarity mixing for time series augmentation

doi:10.1007/s11704-024-40110-9

Front. Comput. Sci.

2024, Vol. 18

Issue (6) : 186352 https://doi.org/10.1007/s11704-024-40110-9

Artificial Intelligence

ISM: intra-class similarity mixing for time series augmentation

Pin LIU^1,², Rui WANG²(

), Yongqiang HE¹, Yuzhu WANG¹

¹. School of Information Engineering, China University of Geosciences, Beijing 100083, China
². State Key Lab of Software Development Environment, Beihang University, Beijing 100191, China

Download: PDF(1264 KB) HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Corresponding Author(s): Rui WANG

Just Accepted Date: 13 May 2024 Issue Date: 07 June 2024

Cite this article:

Pin LIU,Rui WANG,Yongqiang HE, et al. ISM: intra-class similarity mixing for time series augmentation[J]. Front. Comput. Sci., 2024, 18(6): 186352.

URL:

https://academic.hep.com.cn/fcs/EN/10.1007/s11704-024-40110-9
https://academic.hep.com.cn/fcs/EN/Y2024/V18/I6/186352

Fig.1 The overall pipeline of ISM performs time series augmentation from two original time series to augmented time series

Tab.1 Classification performance (Mean Acc.%). F and R represent the backbones of the FCN and ResNet, respectively

Tab.2 The time consumption (minute) on the typical datasets

Fig.2 The influence of batch size on ISM’s performance. fcn_sm and resnet_sm indicate the FCN and ResNet classification models when using ISM for data augmentation, respectively. (a) FordA; (b) WormsTC.

1	W Buntine . Machine learning after the deep learning revolution. Frontiers of Computer Science, 2020, 14( 6): 146320
2	F, Zhang K H, Ngo S, Yang A Nosratinia . Transmit correlation diversity: Generalization, new techniques, and improved bounds. IEEE Transactions on Information Theory, 2022, 68( 6): 3841–3869
3	K, Wei T, Li F, Huang J, Chen Z He . Cancer classification with data augmentation based on generative adversarial networks. Frontiers of Computer Science, 2022, 16( 2): 162601
4	M, Ragab E, Eldele W L, Tan C S, Foo Z, Chen M, Wu C K, Kwoh X Li . ADATIME: a benchmarking suite for domain adaptation on time series data. ACM Transactions on Knowledge Discovery from Data, 2023, 17( 8): 106
5	B, Kim M A, Alawami E, Kim S, Oh J, Park H Kim . A comparative study of time series anomaly detection models for industrial control systems. Sensors, 2023, 23( 3): 1310
6	G, Li J J Jung . Deep learning for anomaly detection in multivariate time series: approaches, applications, and challenges. Information Fusion, 2023, 91: 93–102
7	S, Yun D, Han S, Chun S J, Oh Y, Yoo J Choe . CutMix: regularization strategy to train strong classifiers with localizable features. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. 2019, 6022−6031
8	D, Bellos M, Basham T, Pridmore A P French . A convolutional neural network for fast upsampling of undersampled tomograms in X-ray CT time-series using a representative highly sampled tomogram. Journal of Synchrotron Radiation, 2019, 26( 3): 839–853
9	B K, Iwana S Uchida . Time series data augmentation for neural networks by time warping with a discriminative teacher. In: Proceedings of the 25th International Conference on Pattern Recognition (ICPR). 2021, 3558−3565

[1]

Download

Viewed

Full text

Abstract

Cited

Shared

Discussed