Classification and Regression Trees and MLP Neural Network to Classify Water Quality of Canals in Bangkok, Thailand

Abstract

Water quality is one of the major concerns of countries around the world. This study endeavors to automatically classify water quality. The water quality classes are evaluated using 6 factor indices. These factors are pH value (pH), Dissolved Oxygen (DO), Biochemical Oxygen Demand (BOD), Nitrate Nitrogen (NO 3 N), Ammonia Nitrogen (NH 3 N) and Total Coliform (T-Coliform). The methodology involves applying data mining techniques using classification and regression tree (CART) compared with multilayer perceptron (MLP) neural network models. The data consisted of 288 canals in Bangkok, Thailand. The data is obtained from the Department of Drainage and Sewerage Bangkok Metropolitan Administration during 2003-2007. The results of classification trees perform better than multilayer perceptron neural network. Classification trees exhibit a high accuracy rate at 99.96% in classifying the water quality of canals in Bangkok. Subsequently, this encouraging result could be applied with plan and management source of water quality.