Time series generated by end users, edge devices, and different wearables are mostly unlabelled. We propose a method to auto-generate labels of unlabelled time series, exploiting very few representative labelled time series. Our method is based on representation learning using auto-encoded compact sequence, with a choice of best distance measure. It performs self-correction in iterations by learning latent structure as well as synthetically boosting representative time series using variational auto-encoder to improve the quality of labels.
We have experimented with UCR and UCI archives, public real-world univariate and multivariate time series taken from different application domains. Experimental results demonstrate that the proposed method is very close to the performance achieved by fully supervised classification. It not only produces close-to-benchmark results but outperforms the benchmark performance in some cases.
Research area: Embedded devices and intelligent systems
Authors: Soma Bandyopadhyay, Anish Datta, and Arpan Pal
Conference/event: International Joint Conference on Artificial Intelligence
Conference date: July, 2021