LVQ and PNN Based Classifier Design for BOD of water

This research work is an attempt to estimate the level of water pollution on the basis of available secondary data. The MATLAB is used to develop neural networks for classification of the water pollution levels. Basically the parameter chosen to evaluate water pollution is the BOD (Biological Oxygen Demand). The neural networks are trained to classify the given pH and temperature inputs into 2 classes viz. Class 1 (BOD < 1) and Class 2
$(\text{BOD} \geq 1)$
. The classification accuracy of developed classifiers is also compared.