Using Data Analysis Methods for Predicting the Concentration of Toxic Elements in Soil

Lyazat Naizabayeva, Gulnara Zakirova

2023

Closed access Metadata (CSL-JSON)
DOI 10.1109/idaacs58523.2023.10348723
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

This article is aimed at the selection of machine learning algorithms for the perception of elements in the composition of the soil. In our study, machine learning aims to track changes in the toxicity of contaminated soils in order to predict the level of human health hazard of a contaminated soil condition. For this, the following tasks were set: 1) collection and analysis of data on the types of soil toxicity; 2) study of a mathematical model of toxicity to assess the level of resistance; 3) testing the concentration of toxic elements in soils using four machine learning algorithms. Four machine learning algorithms were tested in the article to predict the content of heavy metals in the soil. The results show that the use of machine learning algorithms allows achieving a high degree of prediction of harmful elements concentrations in soils, which can be useful for making decisions on managing pollution sites. Further improvement of data accuracy and selection of the model and algorithm can also increase the accuracy of prediction. As a result of forecasting using the K-Nearest Neighbors algorithm, the accuracy of the model was achieved at the level of 68%, and using the Decision Tree - 69%. The results of the experiment make it possible to predict the measures of danger to human health from the state of contaminated soil. This will help reduce the risks of the relationship between soil pollution and human health. The obtained results can be useful for specialists in environmental protection, land resource management, and decision-making.