Explainable Time-Series Prediction Using a Residual Network and Gradient-Based Methodsopen access
- Authors
- Choi, H.; Jung, C.; Kang, T.; Kim, H.J.; Kwak, I.
- Issue Date
- Oct-2022
- Publisher
- Institute of Electrical and Electronics Engineers Inc.
- Keywords
- Convolutional neural networks; Data mining; Data models; Feature extraction; Logic gates; Neural networks; Predictive models; Recurrent neural networks; Time series analysis; Time series analysis
- Citation
- IEEE Access, v.10, pp 108469 - 108482
- Pages
- 14
- Journal Title
- IEEE Access
- Volume
- 10
- Start Page
- 108469
- End Page
- 108482
- URI
- https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/61195
- DOI
- 10.1109/ACCESS.2022.3213926
- ISSN
- 2169-3536
- Abstract
- Researchers are employing deep learning (DL) in many fields, and the scope of its application is expanding. However, because understanding the rationale and validity of DL decisions is difficult, a DL model is occasionally called a black-box model. Here, we focus on a DL-based explainable time-series prediction model. We propose a model based on long short-term memory (LSTM) followed by a convolutional neural network (CNN) with a residual connection, referred to as the LSTM-resCNN. In comparison to one-dimensional CNN, bidirectional LSTM, CNN-LSTM, LSTM-CNN, and MTEX-CNN models, the proposed LSTM-resCNN performs best on the three datasets of fine dust (PM2.5), bike-sharing, and bitcoin. Additionally, we tested with Grad-CAM, Integrated Gradients, and Gradients, three gradient-based approaches for the model explainability. These gradient-based techniques combined very well with the LSTM-resCNN model. Variables and time lags that considerably influence the explainable time-series prediction can be identified and visualized using gradients and integrated gradients. Author
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - College of Business & Economics > Department of Applied Statistics > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/61195)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.