Authors :
Rinkal Chauhan; Bela Shrimali
Volume/Issue :
Volume 7 - 2022, Issue 12 - December
Scribd :
https://bit.ly/3EUDlzi
DOI :
https://doi.org/10.5281/zenodo.7698445
Abstract :
Water sources are often polluted due to
human intervention. Water pollution may be classified
according to its quality, which is governed by factors
such as pH, turbidity, the conductivity of dissolved
oxygen (DO), nitrate, temperature, and biological
oxygen demand (BOD). This research compares water
quality categorization methods that use machine
learning techniques, namely SVM, Decision Tree and
Naïve Bayes. The characteristics considered to
determine the quality of water are:
pH, DO, BOD and electrical conductivity.
Classification models are formed based on the
numerical water quality index (WAWQI). After
evaluating the results, With an accuracy of 98.50%, the
decision tree approach was determined to be a better
categorization model.
Keywords :
Classification model; decision tree; support vector machine; naïve bayes; water quality index.
Water sources are often polluted due to
human intervention. Water pollution may be classified
according to its quality, which is governed by factors
such as pH, turbidity, the conductivity of dissolved
oxygen (DO), nitrate, temperature, and biological
oxygen demand (BOD). This research compares water
quality categorization methods that use machine
learning techniques, namely SVM, Decision Tree and
Naïve Bayes. The characteristics considered to
determine the quality of water are:
pH, DO, BOD and electrical conductivity.
Classification models are formed based on the
numerical water quality index (WAWQI). After
evaluating the results, With an accuracy of 98.50%, the
decision tree approach was determined to be a better
categorization model.
Keywords :
Classification model; decision tree; support vector machine; naïve bayes; water quality index.
