Water Quality Prediction Using Catboost Classifier Algorithm

Abstract

Human existence depends on water, so its purity must be assessed for improved consumption. The goal is to use a cat-boost algorithm to forecast the water quality findings as accurately as possible. It is anticipated that improving water quality will lessen health-related issues. By gathering a variety of data, including variable detection and analysis methods like univariate, bivariate, and multivariate analysis, the study of the water quality is carried out. The study compares and discusses how different algorithms function. Data gathering is the first step in the process, during which historical information about the water condition is gathered. With both dependent and independent factors, data analysis is performed. The main objective is to identify the physicochemical characteristics of water, including temperature, pH, EC, hardness, chlorides, alkalinity, phosphate, and sulfate in water samples taken from various monitoring sites. To forecast the trend of data and the quality of the water, the appropriate algorithm is used. The structure has logins for users and government officers, with the latter having access to submit complaints to officials. In order for the police to quickly locate the user, the user is also permitted to provide information about the water purity in addition to their residential data.