<![CDATA[Illegal Gold Mining (PETI) activities in Kuantan Singingi Regency have caused river water pollution and posed serious threats to environmental sustainability and public health. Conventional water quality monitoring methods still have limitations because they rely on periodic laboratory testing and are unable to provide rapid predictive results. Therefore, this study developed a Machine Learning-based prediction system to analyze river pollution levels caused by illegal gold mining activities. The study utilized water quality parameters consisting of pH, temperature, Total Suspended Solid, Dissolved Oxygen, Biological Oxygen Demand, Chemical Oxygen Demand, and mercury concentration. The dataset was processed through several preprocessing stages, including data cleaning, normalization, feature selection, and data splitting. Several Machine Learning algorithms, namely Random Forest, Support Vector Machine, and Artificial Neural Network, were implemented and compared to determine the best prediction model. The results showed that the Random Forest algorithm achieved the best performance with high accuracy and stable classification results. Furthermore, the developed model was integrated into a web-based system equipped with pollution visualization features, river information, and a Geographic Information System. The system is expected to support environmental monitoring and assist decision-making in river pollution management in Kuantan Singingi Regency.]]>
