Developing a comprehensive evaluation model of variety adaptability based on machine learning method
文献类型： 外文期刊
作者： Han, Yanyun 1 ; Wang, Kaiyi 1 ; Zhang, Qi 1 ; Yang, Feng 1 ; Pan, Shouhui 1 ; Liu, Zhongqiang 1 ; Zhang, Qiusi 1 ;
作者机构： 1.Beijing Acad Agr & Forestry Sci, Informat Technol Res Ctr, Beijing, Peoples R China
2.Natl Engn Res Ctr Informat Technol Agr, Beijing, Peoples R China
关键词： Maize variety adaptability evaluation; Variety adaptability comprehensive evaluation index; Entropy weight method; Machine learning method
期刊名称：FIELD CROPS RESEARCH （ 影响因子：5.8； 五年影响因子：6.9 ）
ISSN： 0378-4290
年卷期： 2024 年 306 卷
页码：
收录情况： SCI
摘要： Context or problem: A comprehensive evaluation of the adaptability of maize varieties is very important to accurately promote new varieties and reduce the risk of using them. However, challenges are faced when accurately promoting new varieties owing to the lack of a comprehensive model to evaluate the adaptability of a variety for a specific area, such as a district or county.Objective or research question: The purpose of this study was to construct a new maize (Zea mays) Variety Adaptability Comprehensive Evaluation Index (VACEI) by combining nine agronomic traits; including the lodging rate, stalk rot (Fusarium graminearum) resistance, ear rot (Fusarium graminearum) resistance, Curvularia leaf spot (Curvularia lunata) resistance, southern leaf blight (Bipolaris maydis) resistance, common smut (Ustilago maydis) resistance, southern corn rust (Puccinia polysora) resistance, growth period, and yield; that utilized a machine learning method to construct a prediction model for the VACE and estimate the adaptability of maize varieties on a district or county scale.Methods: The entropy weight method was used to calculate a new maize VACEI by combining nine agronomic traits. A VACEI prediction model that utilized machine learning algorithms was proposed for the application of a variety and its promotion for a locality, such as a district or county. The variety breeding value, seven meteorological factors, such as the effective accumulated temperature, amount of precipitation, duration of total sunshine, daily maximum temperature, daily minimum temperature, daily maximum wind speed, and daily relative humidity; and five soil features, including the soil pH, soil organic matter, soil total nitrogen content, soil available phosphorus, and soil available potassium; were selected as the input parameters of the model. VACEI was set as the output parameter of the model. Gaussian process regression, support vector regression, random forest, and artificial neural network algorithms were used to construct the model.Results: The random forest model outperformed the other algorithms in predicting the VACEI of maize varieties by achieving a maximum coefficient of determination of 0.95 and minimal root mean square error of 6.6316, mean absolute error of 4.6787, and mean square error of 43.978.Conclusions: The constructed model can be used to effectively predict the VACEI of specific varieties to support good varieties in the specific area, such as the selection of a district or county. It can help farmers to objectively select varieties suitable for the local environment and support the precise promotion and planting of cultivars.
- 相关文献
作者其他论文 更多>>
-
A Study of Maize Genotype-Environment Interaction Based on Deep K-Means Clustering Neural Network
作者：Bai, Longpeng;Bai, Longpeng;Wang, Kaiyi;Zhang, Qiusi;Zhang, Qi;Wang, Xiaofeng;Pan, Shouhui;He, Xuliang;Li, Ran;Zhang, Dongfeng;Han, Yanyun;Wang, Kaiyi;Pan, Shouhui;Zhang, Liyang;He, Xuliang;Li, Ran
关键词：small ecological region delineation; deep k-means clustering neural network; genotype by environment interaction
-
Hybrid Deep Learning Approaches for Improved Genomic Prediction in Crop Breeding
作者：Li, Ran;Zhang, Dongfeng;Han, Yanyun;Liu, Zhongqiang;Zhang, Qiusi;Zhang, Qi;Wang, Xiaofeng;Pan, Shouhui;Sun, Jiahao;Wang, Kaiyi;Li, Ran;Pan, Shouhui;Sun, Jiahao;Wang, Kaiyi
关键词：CNN; hybrid models; LSTM; phenotypic prediction; ResNet
-
Multi-Channel Graph Convolutional Network for Evaluating Innovation Capability Toward Sustainable Seed Enterprises
作者：Tang, Shanshan;Wang, Kaiyi;Wang, Kaiyi;Yang, Feng;Pan, Shouhui;Wang, Kaiyi;Yang, Feng;Pan, Shouhui
关键词：innovation capacity evaluation; graph neural network; multi-channel; seed enterprises sustainability
-
VMGP: A unified variational auto-encoder based multi-task model for multi-phenotype, multi-environment, and cross-population genomic selection in plants
作者：Zhao, Xiangyu;Li, Jinlong;Zhang, Dongfeng;Zhang, Qiusi;Liu, Zhongqiang;Wang, Kaiyi;Sun, Fuzhen;Tan, Changwei;Ma, Hongxiang;Zhao, Xiangyu;Wang, Kaiyi;Zhao, Xiangyu;Zhang, Dongfeng;Liu, Zhongqiang;Wang, Kaiyi
关键词：Genomic selection; Variational auto-encoder; Multi-task; Deep learning; Genomic prediction
-
The PeachSNP170K array facilitates insights into a large-scale population relatedness and genetic impacts on citrate content and flowering time
作者：Xu, Yaoguang;Yu, Yang;Qi, Xinpeng;Zhang, Qi;Guan, Jiantao;Zhang, Zhengquan;Wei, Jianhua;Xie, Hua
关键词：
-
Multi-view hypergraph networks incorporating interpretability analysis for predicting lodging in corn varieties
作者：Wang, Kaiyi;Yang, Feng;Zhao, Xiangyu;Liu, Zhongqiang;Zhang, Qiusi;Li, Jinlong;Zhang, Dongfeng;Bai, Wenqin;Wang, Shun;Zhang, Yong;Wang, Kaiyi;Yang, Feng;Zhao, Xiangyu;Liu, Zhongqiang;Zhang, Qiusi;Li, Jinlong;Zhang, Dongfeng
关键词：Corn lodging classification; Multi-view hypergraph network; Graph interpretability
-
Prediction of maize cultivar yield based on machine learning algorithms for precise promotion and planting
作者：Han, Yanyun;Wang, Kaiyi;Yang, Feng;Pan, Shouhui;Liu, Zhongqiang;Zhang, Qiusi;Zhang, Qi;Han, Yanyun;Wang, Kaiyi;Yang, Feng;Pan, Shouhui;Liu, Zhongqiang;Zhang, Qiusi;Zhang, Qi
关键词：Prediction of maize cultivar yield; Machine learning; Random forest; Levenberg - Marquardt neural network; Multilayer perceptron neural network; Assessment of varieties
