Publication

Abstract : Water is a fundamental necessity to life. However, there is a growing scarcity of water and an alarming degradation of quality due to factors such as population growth and urbanization. The deterioration of water is a problem that has harmful impacts on existing and subsequent generations. Water quality control and enhancement is another crucial field of investigation. As the fields of artificial intelligence and machine learning continue to advance, these need to be applied to water quality evaluation. The impact of the subjectivity of water quality index models is employed by Principal component analysis as it reduces the dimensions of the parameters. Thus this work concentrates on the assessment of the performance difference of the classification models of water using feature selection methods including Principal Component Analysis, Sparse Principal Component Analysis, Incremental Principal Component Analysis, and MiniBatch Sparse Principal Component Analysis as Principal Component Analysis reduce The evaluation metrics used to choose the best model among all are accuracy and computational time. The results conveyed that the optimal model is XGboost with Sparse Principal Component Analysis, which has an accuracy of 0.86 and a calculation time of 11.62 s compared to the other models.