Waveel: a wavelet-based ensemble learning framework for time series forecasting
Abstract With the booming development of the Internet of Things, billions of smart devices and sensors have generated massive amounts of time series data. Accurate time series forecasting (TSF) can help to uncover deep underlying patterns in the data. However, most existing TSF models focus on tempo...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2025-07-01
|
| Series: | EURASIP Journal on Wireless Communications and Networking |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13638-025-02479-4 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract With the booming development of the Internet of Things, billions of smart devices and sensors have generated massive amounts of time series data. Accurate time series forecasting (TSF) can help to uncover deep underlying patterns in the data. However, most existing TSF models focus on temporal features and ignore the rich frequency information in time series data. Additionally, due to the non-stationarity of real-world data, previous works face challenges of adapting to diverse data distributions across domains, leading to difficulties in ensuring consistent performance and generalizability in varying scenarios. To address these issues, we propose a wavelet-based ensemble learning framework WaveEL for TSF. WaveEL converts time series data into wavelet domain coefficients while capturing both temporal and frequency domain features. We further integrate multiple TSF models and propose a weight decision maker for determining the weight of model combinations in TSF tasks. We conduct extensive experiments on seven real-world datasets and the results show that our model exhibits superior prediction performance. |
|---|---|
| ISSN: | 1687-1499 |