SVM Parameter Optimization with PSO for Sensor-Based Water Quality Classification and Monitoring Dashboard

Water quality monitoring is essential for environmental sustainability and public health protection. Conventional laboratory-based testing is often time-consuming and unsuitable for real-time monitoring systems. The development of sensor-based Internet of Things (IoT) technology enables continuous acquisition of water quality parameters such as pH, temperature, turbidity, and Total Dissolved Solids (TDS). However, accurate classification of water quality from multi-parameter sensor data remains a challenge due to non-linear data characteristics and the sensitivity of machine learning models to parameter selection. This study aims to optimize the parameters of Support Vector Machine (SVM) using Particle Swarm Optimization (PSO) for sensor-based water quality classification and to integrate the optimized model into a real-time monitoring dashboard. A quantitative experimental approach was employed by comparing the performance of standard SVM and PSO-optimized SVM models. The dataset consisted of sensor measurements collected over 30 days and was evaluated using accuracy, precision, recall, and F1-score metrics. The results show that parameter optimization significantly improves classification performance and enhances the model’s ability to detect critical water quality conditions. The optimized SVM model was successfully integrated into a web-based dashboard capable of real-time monitoring and classification. This study demonstrates that combining metaheuristic optimization with machine learning provides an effective and practical solution for intelligent water quality monitoring systems