Ensembles of Gradient Boosting Recurrent Neural Network for Time Series Data Prediction
Ensemble deep learning can combine strengths of neural network and ensemble learning, which gradually becomes a new emerging research direction. However, the existing methods either lack theoretical support or demand large integrated models. In this paper, Ensembles of Gradient Boosting Recurrent Ne...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9438681/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839619823450980352 |
|---|---|
| author | Shiqing Sang Fangfang Qu Pengcheng Nie |
| author_facet | Shiqing Sang Fangfang Qu Pengcheng Nie |
| author_sort | Shiqing Sang |
| collection | DOAJ |
| description | Ensemble deep learning can combine strengths of neural network and ensemble learning, which gradually becomes a new emerging research direction. However, the existing methods either lack theoretical support or demand large integrated models. In this paper, Ensembles of Gradient Boosting Recurrent Neural Network (EGB-RNN) is proposed, which combines the gradient boosting ensemble framework with three types of recurrent neural network models, namely Minimal Gated Unit (MGU), Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM). RNN model is used as base learner to integrate an ensemble learner, through the way of gradient boosting. Meanwhile, for ensuring the ensemble model fit data better, Step Iteration Algorithm is designed to find an appropriate learning rate before models being integrated. The proposed method is tested on four time-series datasets. Experimental results demonstrate that with the number of integration increasing, the performance of three types of EGB-RNN models tend to converge and the best EGB-RNN model and the best degree of ensemble vary with data sets. It is also shown in statistical results that the designed EGB-RNN models perform better than six baselines. |
| format | Article |
| id | doaj-art-54d93d38f41748d6a27288cfae5d9eb8 |
| institution | Matheson Library |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-54d93d38f41748d6a27288cfae5d9eb82025-07-22T23:00:59ZengIEEEIEEE Access2169-35362025-01-011312233112234010.1109/ACCESS.2021.30825199438681Ensembles of Gradient Boosting Recurrent Neural Network for Time Series Data PredictionShiqing Sang0https://orcid.org/0000-0002-7267-8326Fangfang Qu1https://orcid.org/0000-0002-5748-8372Pengcheng Nie2https://orcid.org/0000-0002-2298-1474Jiaxing Vocational and Technical College, Jiaxing, ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, ChinaEnsemble deep learning can combine strengths of neural network and ensemble learning, which gradually becomes a new emerging research direction. However, the existing methods either lack theoretical support or demand large integrated models. In this paper, Ensembles of Gradient Boosting Recurrent Neural Network (EGB-RNN) is proposed, which combines the gradient boosting ensemble framework with three types of recurrent neural network models, namely Minimal Gated Unit (MGU), Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM). RNN model is used as base learner to integrate an ensemble learner, through the way of gradient boosting. Meanwhile, for ensuring the ensemble model fit data better, Step Iteration Algorithm is designed to find an appropriate learning rate before models being integrated. The proposed method is tested on four time-series datasets. Experimental results demonstrate that with the number of integration increasing, the performance of three types of EGB-RNN models tend to converge and the best EGB-RNN model and the best degree of ensemble vary with data sets. It is also shown in statistical results that the designed EGB-RNN models perform better than six baselines.https://ieeexplore.ieee.org/document/9438681/Gradient boostingLSTMGRUMGUensemble learningtime series data prediction |
| spellingShingle | Shiqing Sang Fangfang Qu Pengcheng Nie Ensembles of Gradient Boosting Recurrent Neural Network for Time Series Data Prediction IEEE Access Gradient boosting LSTM GRU MGU ensemble learning time series data prediction |
| title | Ensembles of Gradient Boosting Recurrent Neural Network for Time Series Data Prediction |
| title_full | Ensembles of Gradient Boosting Recurrent Neural Network for Time Series Data Prediction |
| title_fullStr | Ensembles of Gradient Boosting Recurrent Neural Network for Time Series Data Prediction |
| title_full_unstemmed | Ensembles of Gradient Boosting Recurrent Neural Network for Time Series Data Prediction |
| title_short | Ensembles of Gradient Boosting Recurrent Neural Network for Time Series Data Prediction |
| title_sort | ensembles of gradient boosting recurrent neural network for time series data prediction |
| topic | Gradient boosting LSTM GRU MGU ensemble learning time series data prediction |
| url | https://ieeexplore.ieee.org/document/9438681/ |
| work_keys_str_mv | AT shiqingsang ensemblesofgradientboostingrecurrentneuralnetworkfortimeseriesdataprediction AT fangfangqu ensemblesofgradientboostingrecurrentneuralnetworkfortimeseriesdataprediction AT pengchengnie ensemblesofgradientboostingrecurrentneuralnetworkfortimeseriesdataprediction |