首页|基于LSBoost模型的农业产业结构分析——以广东省为例

基于LSBoost模型的农业产业结构分析——以广东省为例

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
通过构建LSBoost(最小二乘提升)模型,以广东省农业历年林业、牧业、渔业、粮食种植业各产业产值增长率为自变量、农业总产值增长率为因变量构建回归模型进行分析.结果表明,除粮食产量外,其余各产业均保持正增长,其中渔业产值最高达到年均11.55％.从LSBoost模型的相关性参数分析可以得出渔业对广东省农业经济增长贡献的占比最大,约为34％;而林业、牧业、粮食种植业则分别占比23.06％、33.8％、9.15％.说明渔业与牧业为广东省农业产业的支柱产业.通过与其他4种模型进行比较,LSBoost模型的5项相关评价指标均优于其他模型.实验验证了LSBoost对于广东省农业产业产值增长率数据的拟合以及特征相关性分析的高精度以及鲁棒性,为农业产业产值结构分析预测提供了 一个较好的评价以及回归拟合模型.
Structural Analysis of Agricultural Industry Based on LSBoost Model:Taking Guangdong Province as an Example
The growth rates of the agricultural industries in Guangdong Province is analyzed,including forestry,animal husbandry,fisheries,and grain planting,by constructing an LSBoost(least square boost)model with the growth rates of the industries as independent variables and the total agricultural output value growth rate as the dependent variable.The results indicate that except for grain production,all other industries maintained positive annual growth,with the fisheries industry reaching the highest annual average growth rate of 11.55％.Analysis of the correlation parameters of the LSBoost model reveals that the fisheries industry contributes the largest proportion to the agricultural economic growth of Guangdong Province,accounting for approximately 34％.Meanwhile,forestry,animal husbandry,and grain planting account for proportions of 23.06％,33.8％,and 9.15％respectively,indicating that fisheries and animal husbandry are the backbone industries of the agricultural sector in Guangdong Province.Compared with four other models,the LSBoost model outperforms in all five related evaluation indicators.This experiment verifies the high accuracy and robustness of LSBoost in fitting the data on the growth rates of agricultural industry output in Guangdong Province and in analyzing feature correlations,providing a better evaluation and regression fitting model for predicting the structure of agricultural industry output values.

agricultural industry growth rateLSBoost(least square boost)modelregression analysischaracteristic analysis

陈雨秋、梁凯豪

展开 >

仲恺农业工程学院经贸学院,广州 510000

仲恺农业工程学院数学系,广州 510000

农业产业增长率 LSBoost(最小二乘提升)模型 回归分析 特征性分析

广东省自然科学基金广东省自然科学基金广东省科技计划项目广东省科技计划项目广东省教育厅项目省级乡村振兴战略专项大学生创新创业训练计划项目广东省农产品保鲜物流共性技术创新团队项目广州市科技计划项目广东省高校特色创新项目(自然科学)粤港澳大湾区农产品数字化物流研究中心项目

2018A03031301362017A0303106502015A0707040592015A0304020082020KZDZX1120粤农计[2018]51号2022113470362023KJ1452017040301312018KTSCX094

2024

科技和产业
中国技术经济学会

科技和产业

影响因子：0.361
ISSN：1671-1807
年,卷(期)：2024.24(13)