Random Forest Algorithm for Mitigating Floods and Ensuring Clean Water: IoT-Based Monitoring, Harvesting and Purification in Flood-Prone India

The UN SDG 6 strives to provide access to clean drinking water worldwide. India is currently facing an increasing vulnerability to natural disasters, particularly due to the rising frequency of floods. This article addresses the pressing issues of floods and water scarcity in India by proposing a flood monitoring and water purification system based on Internet of Things (IoT). The research evaluates the shortcomings of existing flood warning systems and underscores the critical importance of timely public alerts. A substantial portion of India's population has limited access to global water resources, leading to a decrease in the annual per capita water availability. The proposed solution involves an IoT framework equipped with various sensors for flood monitoring, including those that measure total dissolved solids (TDS), temperature, humidity, and water levels. Additionally, traditional charcoal-based water purification methods are explored to enhance water quality. Utilizing TDS readings, the study employs supervised machine learning techniques, specifically Random Forest and Support Vector Machine (SVM), to predict the suitability of water for drinking. The empirical results highlight the superior performance of Random Forest, achieving an impressive accuracy rate of 91.76 percent.