In given propose paper we have worked on the water quality of river ganga, not only for management of water resources, but also for the prevention of water pollution, the water quality forecast has a more practical significance. To evolvesuitableideals for the water quality (WQ) insidethe water physiquesobtainingcontaminant samples & then to confirm that these standards are encountered, this is the environmentalWQ management program have goal. In the realistic standard setting, the institutional capacity of the basin’s water science, environmental, & the land usagesituations, possibleusages of getting water bodies, &the determination & implementation of WQ ideals has been kept in mind. In this paper, an efficient Machine learning algorithmwas modeled. A feed forward error back propagation neural network is implemented with different training functionsnamely trainlm(LevenbergMarquardtbackpropagation), trainb(Batch training with weight & bias learning rules), trainr(Random order incremental training w/learning functions) & trainbr(Bayesian regularization). Five sampling stations along Ganga River stretch were selected from DEVPRAYAG-to-ROORKEE city inside the Uttarakhand state of the India.These states are Bihar, Uttarakhand, Delhi, UP& West Bengal. The hill rivers of the Uttarakhand are Alkananda, Bhagirathi, Mandakini.We have used the above given training functions at different-different learning rate (i.e., 0.01,0.03,0.05,0.07&0.09) to measure classification rate& use the mean square method to measure the performance of the model. Results indicate that the proposed algorithm gives best estimating model and generate less mean square error (i.e., 0.004)and accuracy is 99.5% at 0.09 learning rate with trainbr training functionin respect to othertraining function.