Journal of Zhejiang University SCIENCE  B

Accepted manuscript available online (unedited version)


Ensemble learning for wheat yield prediction using multi-source and multi-temporal data


Author(s):  Runzhou CAO1, Xuyang SUO1, Xuping FENG1, Yong HE2

Affiliation(s):  1College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China; more

Corresponding email(s):  yhe@zju.edu.cn

Key Words:  Wheat, Machine learning, Yield prediction, Remote sensing


Runzhou CAO1, Xuyang SUO1, Xuping FENG1, Yong HE2. Ensemble learning for wheat yield prediction using multi-source and multi-temporal data[J]. Journal of Zhejiang University Science B,in press.Frontiers of Information Technology & Electronic Engineering,in press.https://doi.org/10.1631/jzus.B2600177

@article{title="Ensemble learning for wheat yield prediction using multi-source and multi-temporal data",
author="Runzhou CAO1, Xuyang SUO1, Xuping FENG1, Yong HE2",
journal="Journal of Zhejiang University Science B",
year="in press",
publisher="Zhejiang University Press & Springer",
doi="https://doi.org/10.1631/jzus.B2600177"
}

%0 Journal Article
%T Ensemble learning for wheat yield prediction using multi-source and multi-temporal data
%A Runzhou CAO1
%A Xuyang SUO1
%A Xuping FENG1
%A Yong HE2
%J Journal of Zhejiang University SCIENCE B
%P
%@ 1673-1581
%D in press
%I Zhejiang University Press & Springer
doi="https://doi.org/10.1631/jzus.B2600177"

TY - JOUR
T1 - Ensemble learning for wheat yield prediction using multi-source and multi-temporal data
A1 - Runzhou CAO1
A1 - Xuyang SUO1
A1 - Xuping FENG1
A1 - Yong HE2
J0 - Journal of Zhejiang University Science B
SP -
EP -
%@ 1673-1581
Y1 - in press
PB - Zhejiang University Press & Springer
ER -
doi="https://doi.org/10.1631/jzus.B2600177"


Abstract: Measuring crop yields for hundreds of wheat germplasm accessions is time-consuming and labor-intensive in plant breeding. Integrating drone-based remote sensing with machine learning techniques has recently emerged as an effective solution for streamlining this process. Despite these advances, selecting an optimal predictive model remains challenging because machine learning algorithms vary in their principles and characteristics. In this study, support vector regression, random forest, extreme gradient boosting, and gradient boosting decision tree served as base learners and were integrated through blending ensemble learning. Multiple blending ensemble models were developed by varying the proportions of training and hold-out sets, and a weighted average was subsequently applied to the resulting models. The proposed model achieved an R2 of 0.889 and an RMSE of 0.470 t/ha, demonstrating its superior performance, its ability to accurately predict yields across multiple wheat varieties, and its effectiveness in alleviating model-selection challenges. Furthermore, the heading stage was identified as the most suitable growth period for accurate multi-variety wheat yield prediction.

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

Open peer comments: Debate/Discuss/Question/Opinion

<1>

Please provide your name, email address and a comment





Full Text:  <88>

CLC number: 

On-line Access: 2026-10-13

Received: 2026-03-16

Revision Accepted: 2026-09-08

Crosschecked: 0000-00-00

Cited: 0

Clicked: 97

Citations:  Bibtex RefMan EndNote GB/T7714

Journal of Zhejiang University-SCIENCE, 38 Zheda Road, Hangzhou 310027, China
Tel: +86-571-87952783; E-mail: cjzhang@zju.edu.cn
Copyright © 2000 - 2026 Journal of Zhejiang University-SCIENCE