A spatio-temporal graph representation model based on shapelets

Jianping Huang; Xudong Zhang; Siqi Shen; Ke Chen; Xiaofeng Liu

doi:10.1117/12.3004537

11 October 2023 A spatio-temporal graph representation model based on shapelets

Jianping Huang, Xudong Zhang, Siqi Shen, Ke Chen, Xiaofeng Liu

Proceedings Volume 12800, Sixth International Conference on Computer Information Science and Application Technology (CISAT 2023); 128004O (2023) https://doi.org/10.1117/12.3004537
Event: 6th International Conference on Computer Information Science and Application Technology (CISAT 2023), 2023, Hangzhou, China

Abstract

Time series is the data type that can be seen everywhere in practical scenarios. With the rapid development of deep learning, such time series models have made great achievements in various fields. However, the lack of interpretability of deep learning models has been one of the most criticized issues. Inspired by recent work on subsequence mining, we conduct time series modeling from the perspective of shapelets and explore the spatial relationship between different time series samples, thereby designing a new spatio-temporal model which focus on downstream binary classification tasks to obtain feature representations of data samples. We conduct experiments on two datasets that contain both time series and graph structures. The results show that our designed spatio-temporal graph model can effectively learn discriminative feature representations of raw data and achieve competitive results in binary classification tasks.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Jianping Huang, Xudong Zhang, Siqi Shen, Ke Chen, and Xiaofeng Liu "A spatio-temporal graph representation model based on shapelets", Proc. SPIE 12800, Sixth International Conference on Computer Information Science and Application Technology (CISAT 2023), 128004O (11 October 2023); https://doi.org/10.1117/12.3004537

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available