浅谈机器学习在土壤重金属监测中的应用进展  

Application Progress of Machine Learning in Soil Heavy Metal Monitoring

作　　者：王柳毅 WANG Liuyi(College of Computer and Information Technology of China Three Gorges University,Yichang 443000,Hubei,China)

机构地区：[1]三峡大学计算机与信息学院,湖北宜昌443000

出　　处：《长江信息通信》2023年第11期63-65,共3页Changjiang Information & Communications

摘　　要：随着我国工业化快速发展,土壤污染速度也随之加快,尤其是土壤重金属污染,土壤中的重金属会通过农产品进入人体,对人类的健康造成威胁。地统计模型基于土壤的空间特性对土壤重金属直接进行插值,但无法真正表征辅助变量对土壤重金属空间分异性的影响,现如今,机器学习在各个领域发挥着越来越重要的作用,而且机器学习算法能够将有效整合辅助变量和土壤重金属数据,因此探讨哪种机器学习算法能最大程度上提高对土壤重金属含量预测的精度具有重要的现实意义。文章对传统的机器学习算法进行梳理,浅谈机器学习在土壤重金属预测中的最新研究并进行相关分析,以期为后续土壤重金属预测相关研究提供参考。With the rapid development of industrialization in China,the speed of soil pollution has also accelerated,especially heavy metal pollution in soil.Heavy metals in soil can enter the human body through agricultural products,posing a threat to human health.Geostatistical models directly interpolate soil heavy metals based on the spatial characteristics of the soil,but can-not truly characterize the impact of auxiliary variables on the spatial heterogeneity of soil heavy metals.Nowadays,machine learning plays an increasingly important role in various fields,and machine learning algorithms can effectively integrate auxili-ary variables and soil heavy metal data,Therefore,exploring which machine learning algorithm can maximize the accuracy of predicting soil heavy metal content has important practical significance.This article reviews traditional machine learning algo-rithms,discusses the latest research and relevant analysis of machine learning in soil heavy metal prediction,in order to provide reference for subsequent research on soil heavy metal prediction.

关 键 词：土壤重金属 地统计模型 机器学习 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]