An Improved Water Quality Prediction Model with Artificial Neural Network and Adaptive Neuro-Fuzzy Inference System Ensemble
| dc.contributor.author | Yafra Khan | |
| dc.date.accessioned | 2026-10-15T03:40:23Z | |
| dc.date.issued | 2018 | |
| dc.description | The deteriorating quality of natural water resources like lakes, streams and estuaries, is one of the most worrisome issues faced by humanity today. The effects of unclean water are far-reaching, impacting every aspect of life. Therefore, management of water resources is crucial to optimise the quality of water to the utmost degree. The effects of water contamination can be tackled if data is analysed and water quality predicted. The use of machine learning techniques has been proven to be promising for the analysis and prediction models for environmental applications in general and water quality applications in particular. However, there is an ample room for improvement regarding accuracy, robustness, usability as well as overall performance of the current water quality analysis and prediction methodologies. It can be efficiently tackled by integration of machine learning techniques, which utilises the strengths of multiple methods and minimises the drawbacks, ensuring an improved prediction process. The primary objective of water quality analysis and prediction process is to determine the overall quality of a water body and to forecast the values of the concentrations of water quality parameters, which influence the overall water quality. The goal of this study is to develop a water quality prediction model with the help of water quality factors using Artificial Neural Network and Adaptive Neuro-Fuzzy Inference System. The process also involves a feature selection technique using ANN, which selects a subset of the most influential parameters, in order to ensure an improved performance. Finally ANN and ANFIS learning algorithms are combined through an ensemble technique to produce a new model. This research uses the water quality historical data of the year of 2015, with 30-minutes time interval. Data is obtained from the United States Geological Survey online resource called National Water Information System. For this study, the data includes the measurements of iv nine parameters which affect and influence water quality whose subset is later acquired through feature selection and is reduced to a five-parameter dataset. For the purpose of evaluating the performance of the model, the performance evaluation measures used are Root Mean-Squared Error and Correlation Coefficient, and are compared with the performance yielded in latest related studies. | |
| dc.identifier.uri | https://scholarhub.unimas.my/handle/123456789/3101 | |
| dc.language.iso | English | |
| dc.publisher | UNIVERSITI MALAYSIA SARAWAK | |
| dc.relation.ispartofseries | Faculty of Computer Science and Information Technology | |
| dc.subject | Water quality prediction, machine learning, artificial neural networks, adaptive neuro-fuzzy inference system | |
| dc.title | An Improved Water Quality Prediction Model with Artificial Neural Network and Adaptive Neuro-Fuzzy Inference System Ensemble | |
| dc.type | Masters |