Combining Fingerprint Classifiers
MCS '00 Proceedings of the First International Workshop on Multiple Classifier Systems
Large-Scale Evaluation of Multimodal Biometric Authentication Using State-of-the-Art Systems
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.00 |
Water is one of the basic requirements of any community, without water supply, human life is not possible. The next most essential requirement is the elimination of waste matters from the water. Anaerobic wastewater treatment differs from traditional aerobic treatment in that no aeration is used. In this paper, cheese-dairy wastewater is taken for the treatment. The ultimate objective of wastewater treatment is the conservation of good, quality water, the most priceless resource. Chemical Oxygen Demand (COD) is an essential test for determining the quality of effluents and wastewaters prior to discharge. COD test predicts the level of oxygen requirement of the effluent and is exploited for monitoring and control of discharges and for assessing treatment plant performance. The entire process in an anaerobic wastewater treatment system for the prediction of COD includes, data collection and pre-processing (Statistical Normalization technique), Feature Selection and Prediction of COD using Back Propagation Neural Network (BPN). During the pre-processing phase several normalization techniques are used. The major objective of this paper is to propose several statistical normalization techniques and to improve the prediction accuracy. BPN is found to be best for the prediction of COD. In order to find the effect of normalization technique in the prediction of COD, experiments were carried out to find the prediction of accuracy of BPN before/after normalization techniques.