Prediction of water quality for Bangkok canals using data mining techniques

Abstract

Water is usually subject to contamination caused by human activities. The main problems being studied are two folds, 1) derivation of an appropriate model that can achieve accurate classification of water, and 2) construction of an efficient model in predicting the amount of Dissolved Oxygen (DO) that affects water quality of Bangkok's canals based on 13 water quality parameters. These parameters are collected from the data obtained from the Department of Drainage and Sewerage, Bangkok Metropolitan Administration during 2003-2011. This research proposes two prediction models. The Classification and Regression Tree (CART) model is established to classify water quality in Bangkok’s canals. The second model is set up to cluster water quality parameters and integrated the clusters to predict DO with the help of an ANN model. Result comparisons by correlation coefficient (R), mean absolute error (MAE), and mean square error (MSE) between the proposed technique and other techniques show higher accuracy than the comparative techniques. Contributions of this work can be applied to better water quality planning and effective treatment.