Document Type : Research Paper

Authors

• Brintha Malar C ¹
• Siva Sankar K ²

¹ Tirunelveli Dakshina Mara Nadar Sangam College,affiliated with Manonmaniam Sundaranar University.

² Noorul Islam Centre for Higher Education

Abstract

A significant source of water pollution, industrial effluents degrade water quality and jeopardize ecological equilibrium. In order to solve this problem, this study suggests an automated system that uses supervised learning techniques to anticipate the best places for wastewater disposal. The method makes use of statistical characteristics including mean, standard deviation, variance, and entropy in addition to two important water quality indicators: chemical oxygen demand (COD) and biological oxygen demand (BOD). We create and compare four classification models: ensemble classification, Extreme Learning Machine (ELM), Support Vector Machine (SVM), and k-Nearest Neighbour (k-NN). Accuracy, sensitivity, specificity, and computing time are all taken into consideration while evaluating performance. Experimental results show that whereas other approaches provide a balance between efficiency and efficacy, ensemble classification offers improved prediction accuracy at the expense of increased time consumption. In order to ensure adherence to pollution control regulations and assist sustainable water management, the suggested framework offers industry a useful decision-support tool for locating appropriate wastewater disposal sites.

Keywords

• Pollution Mitigation
• Sustainability assessment