ShapeWordNet: An Interpretable Shapelet Neural Network for Physiological Signal Classification
This work addresses classification challenges in medical and healthcare applications, offering an interpretable solution for high-dimensional time series with sparse labels, though it appears incremental in its approach.
The authors tackled physiological signal classification by proposing ShapeWordNet, which uses shapelets for interpretable discretization and contrastive learning to address data sparsity, achieving superior performance over state-of-the-art methods on three real-world datasets.
Physiological signals are high-dimensional time series of great practical values in medical and healthcare applications. However, previous works on its classification fail to obtain promising results due to the intractable data characteristics and the severe label sparsity issues. In this paper, we try to address these challenges by proposing a more effective and interpretable scheme tailored for the physiological signal classification task. Specifically, we exploit the time series shapelets to extract prominent local patterns and perform interpretable sequence discretization to distill the whole-series information. By doing so, the long and continuous raw signals are compressed into short and discrete token sequences, where both local patterns and global contexts are well preserved. Moreover, to alleviate the label sparsity issue, a multi-scale transformation strategy is adaptively designed to augment data and a cross-scale contrastive learning mechanism is accordingly devised to guide the model training. We name our method as ShapeWordNet and conduct extensive experiments on three real-world datasets to investigate its effectiveness. Comparative results show that our proposed scheme remarkably outperforms four categories of cutting-edge approaches. Visualization analysis further witnesses the good interpretability of the sequence discretization idea based on shapelets.