现代农业科技Issue(12)：163-166,4.DOI:10.3969/j.issn.1007-5739.2025.12.036
基于多种机器学习模型的烟叶产情预测研究
Research on Tobacco Production Prediction Based on Multiple Machine Learning Models
伍超祥1
作者信息
- 1. 四川省烟草公司宜宾市公司,四川 宜宾 644400
- 折叠
摘要
Abstract
Taking Xingwen County in Yibin City as the research area,based on the data of meteorology,tobacco yield and average price data of Xingwen County from 2014 to 2023,we selected five machine learning models including random forest,decision tree,extreme random tree,gradient boosting decision tree and CatBoost to analyze the tobacco yield data.The learning and prediction effects of multiple machine learning models were compared and analyzed to find the best tobacco yield prediction model under complex climate conditions.The results showed that the correlation coefficient of the gradient boosting decision tree reached 0.75,the mean absolute error was 20.81,the root mean square error was 20.94,and the prediction accuracy was 93.50%,which had certain guiding significance for the precision and fine management of tobacco leaf production.关键词
烟叶/产情预测/气象要素/机器学习模型Key words
tobacco leaf/production forecast/meteorological element/machine learning model引用本文复制引用
伍超祥..基于多种机器学习模型的烟叶产情预测研究[J].现代农业科技,2025,(12):163-166,4.
