Analysis of Water Quality in Cauvery River by Using PCA with Various Classifications Technique

Water quality assessment is a complex endeavour that involves identifying pollutants in water resources. The importance of this process lies in its objective to evaluate water quality for human use. Machines and deep learning algorithms play a vital role in this evaluation. In this context, feature selection methods were employed to identify critical factors that ensure optimal accuracy. Subsequently, these selected features were used as input for various classifiers for classification purposes. The Cauvery River dataset, obtained from the Tamil Nadu Pollution Control Board, was utilized to assess the performance of the proposed approach. This research was implemented using Python as the programming language. The performance of the PCA-RF model was evaluated using various metrics, including an accuracy of 0.96, precision of 0.97, recall of 0.94, and an F1-score of 0.95. The results demonstrate that the PCA-RF model outperforms conventional machine learning approaches, achieving a high R-squared score of 0.95.