## Introduction

In this paper, we propose a novel algorithm called ShapeNet,
which embeds shapelet candidates from different lengths into the unified space for shaplets selection.
The network is trained using our cluster-wise triplet loss,
which considers the distance between anchor and multiple positive (negative) samples
and the distance among positive (negative) samples.
Then, we compute representative and diversified final shapelets
rather than directly using all the embeddings for model building
to avoid a large fraction of computing non-discriminative shapelet candidates.
A classical classifier (e.g., SVM) is then adopted.

## The overview of ShapeNet

## Source

We implemented the proposed algorithms in Python.
A well-known benchmark of time series classification datasets, UEA Time Series Classification Archive, has been tested.

Source code can be found in here.

The style of password is “xxxxxxxx”,

all the characters are with lower case of our method name.

## Supplementary

- Guozhong Li, Byron Choi, et al. Supplementary Material: ShapeNet: A Shapelet-Neural Network Approach for Multivariate Time Series Classification, Supplementary.

Do one thing at a time, and do well. :)

## References

- Guozhong Li, Byron Choi, et al. ShapeNet: A Shapelet-Neural Network Approach for Multivariate Time Series Classification, AAAI 2021, poster. (to appear)
- Bostrom, Aaron, and Anthony Bagnall. “A shapelet transform for multivariate time series classification.” arXiv preprint arXiv:1712.06428 (2017).
- Lines, Jason, et al. “A shapelet transform for time series classification.” Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2012.
- …

## Acknowledgement

Thanks the research community for supporting the datasets.

HKBU Database Group