基于粒子群优化随机森林算法的河南省葡萄价格预测研究
Research on grape price forecasting of Henan Province based on particle swarm optimization random forest algorithm
王哲 1张文慧 1付金鹏 1杨进进1
作者信息
- 1. 华北水利水电大学信息工程学院,郑州 450000
- 折叠
摘要
河南省是中国重要的葡萄种植基地之一,葡萄种植面积广泛且产量丰富.因此预测河南省葡萄价格至关重要.为了提高葡萄价格预测准确性和稳定性,提出一种PSO-RF预测模型,该模型用粒子群(PSO)优化随机森林中决策树深度,选择最优预测器个数,进一步得到最优预测器组合.实验结果表明,相比于单一的随机森林预测模型,PSO-RF模型具有更高的预测精度,MAE仅为0.0095,R2达到0.968.因此PSO-RF预测模型可以对河南省葡萄价格进行更准确的预测.
Abstract
Henan Province is one of the important grape planting bases in China,with extensive grape planting area and abun-dant yield.Therefore,it is very important to predict the price of grapes in Henan Province.In order to improve the accuracy and sta-bility of grape price prediction,a PSO-RF prediction model is proposed in this paper.In this model,particle swarm optimization(PSO)is used to optimize the depth of decision tree in random forest,select the optimal number of predictors,and further obtain the optimal combination of predictors.The experimental results show that compared with the single random forest prediction model,the PSO-RF model has higher prediction accuracy,MAE is only 0.0095,R2 is 0.968.Therefore,the PSO-RF prediction model can make a more accurate prediction of grape price in Henan Province.
关键词
葡萄价格预测/决策树/随机森林/粒子群算法Key words
grape price forecast/decision tree/random forest/particle swarm optimization引用本文复制引用
出版年
2024
NETL
NSTL
万方数据