Water Quality Parameter Prediction Using Machine Learning Method

Thumbnail Image
Muhammad Shaiful Bin Abdul Halim
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue
This thesis reports the use of machine learning method in predicting the water quality parameter by using machine learning software. Selected algorithm are used to generates a model and by inserting the input values, the model can produce the output. The accuracy of the model was determined by the correct and appropriate number of neuron numbers, value of Mean Square Error (MSE) and the value of regression (R). The closer the MSE value to zero, the more accurate the model and the closer the R value to one, the higher the correlation between the input and the output. This thesis also shows the use of Neural Network Fitting application in the Matlab software that consists of three types of training algorithm which is Levenberg-Marquardt, Bayesian Regularization and Scaled Conjugate Gradient. Training algorithm are chosen depending on several criteria such as the value and difficulty of the data and also the time to train and produce the model. The development of the model can bring significant boost to the modern water quality parameter prediction method and to the water quality itself. Further study on introducing more challenging input data could be done to maximize the efficiency of the model in the future.
FYP Sem 2 2019/2020
Water , Environment , Nitrite