Evaluation of the Quality of Water Considering SVM and the XGBoost Technique

  • Darshan P Department of Master of Computer Applications Raja Rajeswari College of Engineering
  • T Subburaj Department of Master of Computer Applications Raja Rajeswari College of Engineering
Keywords: Water Quality Prediction, XGBOOST, SVM, Machine Learning.

Abstract

Various contaminants have been threatening aquatic excellence for decades. As a result, forecasting and modelling water quality has become serious to reducing water pollution. This study created a classification process for foreseeing water quality (WQC) classifying. By means of a Sustenance Trajectory Appliance and Extreme Gradient Increasing (XGBoost), the WQC is determined by analysing a dataset’s water quality indicator (WQI), which is derived from seven parameters. The suggested model’s outputs may properly classify in accordance with the qualities of the water. The threesearches of this research showed that the application of X algorithm outscored the SVM the structure which had an accuracy rating of 94% but only 67% specificity. Additionally, the error rate for misclassification for the XGBoost was only 6% as opposed to 33% for the SVM. Additionally, XGBoost regularly outperformed SVM in the 5-fold validation, with an regularaccurateness of 90% as opposed to 64% for SVM.Given its improved presentation, XGBoost is believed to be more effective at classifying the cleanliness of water.

Published
2023-07-01
Statistics
Abstract views: 158 times
PDF downloads: 103 times
Section
Articles