相关实验视频

Updated: Jul 19, 2025

The MPLEx Protocol for Multi-omic Analyses of Soil Samples

The MPLEx Protocol for Multi-omic Analyses of Soil Samples

Published on: May 30, 2018

根据多门混合专家网络,识别潜在污染地点的多重土壤污染

Xiaosong Lu¹, Junyang Du¹, Guoqing Wang¹

¹State Environmental Protection Key Laboratory of Soil Environmental Management and Pollution Control, Nanjing Institute of Environmental Sciences, Ministry of Ecology and Environment, Nanjing 210042, China.

The Science of the total environment

|August 12, 2023

概括

本研究引入了一种新的机器学习方法,用于识别工地土壤污染,包括重金属,挥发性有机化合物 (VOC) 和半挥发性有机化合物 (SVOC). 该方法在复杂的污染识别任务中显著提高了准确性和效率.

科学领域:

环境科学环境科学
数据科学数据科学数据科学
地质化学地质化学

背景情况:

大数据和机器学习越来越多地用于网站土壤污染的识别.
之前的方法在多种污染类型的准确性和效率方面扎.

研究的目的:

开发一种更准确,更有效的方法来识别多个地点的土壤污染物.
确定土壤污染的关键驱动因素.

主要方法:

从6个行业的199个地点收集了环境数据.
开发了使用10个关键索引的土壤污染识别指数系统.
使用多门混合专家 (MMoE) 网络同时识别重金属,VOC和SVOC污染.
使用SHAP框架来确定指数和驱动因素的重要性.

主要成果:

在MMoE模型中,重金属的精度为0.600,VOC的精度为0.783,SVOC的精度为0.850.
对于单个任务,准确度比传统的BP神经网络高0-20%.
确定了诸如原材料组成,企业规模和行业类型等因素的重要性.

结论:

拟议的MMoE方法为地点土壤污染的识别提供了更有效,更准确的解决方案.
了解驱动因素对于智能识别和全球土壤污染风险控制至关重要.

关键词:

驱动因素分析驱动因素分析 SHAP框架的模型可解释性多门混合专家网络网络.多任务学习是多任务学习.网站的土壤污染识别网站的土壤污染.

更多相关视频

Single-throughput Complementary High-resolution Analytical Techniques for Characterizing Complex Natural Organic Matter Mixtures

Single-throughput Complementary High-resolution Analytical Techniques for Characterizing Complex Natural Organic Matter Mixtures

Published on: January 7, 2019

Integrated Field Lysimetry and Porewater Sampling for Evaluation of Chemical Mobility in Soils and Established Vegetation

Integrated Field Lysimetry and Porewater Sampling for Evaluation of Chemical Mobility in Soils and Established Vegetation

Published on: July 4, 2014

相关概念视频

Bioremediation

Bioremediation

Bioremediation is the use of prokaryotes, fungi, or plants to remove pollutants from the environment. This process has been used to remove harmful toxins in groundwater as a byproduct of agricultural run-off and also to clean up oil spills.