Development of Regression Models for Predicting Water Quality Index Based on Dissolved Oxygen for River Pollution Assessment

Water is an essential resource in Malaysia, playing a crucial role in sustaining human life, agriculture, and industry. However, rapid industrialization, urbanization, and development have significantly deteriorated river water quality, posing serious environmental and public health risks. Traditional water quality monitoring methods rely on manual sampling and laboratory analysis and are often time-consuming, labor-intensive, and inefficient. This study aims to overcome these challenges by developing regression-based predictive models to estimate the Water Quality Index (WQI) based on Dissolved Oxygen (DO) measurements. The research utilizes a dataset of 219 river water samples collected between June and November 2023 from the Kaggle database. Statistical validation techniques were applied to assess data distribution and accuracy, including normality tests and error bar plots. Multiple regression techniques were implemented using MATLAB and Python to determine the most effective model. MATLAB’s Linear Regression model demonstrated superior performance among the tested approaches, achieving an R² value of 0.95397 and a Root Mean Square Error (RMSE) of 7.2728. These results highlight the potential of regression models in providing a fast, reliable, and cost-effective method for water quality assessment. By leveraging these predictive techniques, environmental authorities and policymakers can implement timely interventions, ensuring better management and protection of freshwater ecosystems in Malaysia.