Water-Quality-Analysis using Machine Learning

• Published in: 2022 11th International Conference on System Modeling & Advancement in Research Trends (SMART) pp. 13 - 18
• Main Authors: Akshay, Raavi, Tarun, Gadiraju, Kiran, Pinapothini Uday, Devi, K Durga, Vidhyalakshmi, M.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 16.12.2022
• Subjects: Classification algorithms; Data models; Data standardization; Dataset pre-processing; Hyperparameter Tuning; Machine learning algorithms; Performance metrics; Training; Urban areas; Water pollution; Water quality
• ISBN: 9781665487320; 1665487321
• ISSN: 2767-7362
• DOI: 10.1109/SMART55829.2022.10047533

One of the most serious and alarming problems facing humanity is the degradation of natural water resources such as lakes as well as rivers is one of the most serious and vexing problems we are facing today. The long-term effects of polluted water affect all areas of life. Therefore, it is essential to manage water resources if you want tomaximize the quality of your water. If data are examinedand water quality can be predicted, the effects of watercontamination can be dealt with effectively. The purpose of this study is to use machine learning to make a water qualityprediction model based on water quality measurements. Machine learning can be used for building models fromalgorithms with some data gathered from the sick ones. For a better examination of parametric findings, the acquired data will be pre-processed, separated into training and testing parts, and subjected to machine learning classification techniques. Decision tree, Naive Bayes, Random Forest, Support Vector Machine, and K-Nearest Neighbor are some of the classification-type algorithms employed in this work. All the model's performance indicators are calculated, and they change for each model. A technique for improving machine learning model performance metrics is hyperparameter tuning.