A comparison of unsupervised and supervised machine learning algorithms to predict water pollutions

Zamri, Nurnadiah and Pairan, Mohammad Ammar and Azman, Wan Nur Amira Wan and Abas, Siti Sabariah and Abdullah, Lazim and Naim, Syibrah and Tarmudi, Zamali and Gao, Miaomiao (2022) A comparison of unsupervised and supervised machine learning algorithms to predict water pollutions. Procedia Computer Science, 204. pp. 172-179. ISSN 1877-0509 (https://doi.org/10.1016/j.procs.2022.08.021)

[thumbnail of Zamri-etal-PCS-2022-A-comparison-of-unsupervised-and-supervised-machine-learning-algorithms-to-predict-water-pollutions]
Text. Filename: Zamri_etal_PCS_2022_A_comparison_of_unsupervised_and_supervised_machine_learning_algorithms_to_predict_water_pollutions.pdf
Final Published Version
License: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 logo

Download (1MB)| Preview


Clean and safe water is vital for our lives and public health. In recent decades, population growth, agriculture, industries, and climate change have worsened freshwater resource depletion and clean water pollution. Several studies have focused on water pollutions risk simulation and prediction in the presence of pollution hotspots. However, the increase and complexity of big data caused by uncertain water quality parameters led to a new efficient algorithm to trace the most accurate pollution hotspots. Therefore, this study proposes to offer different algorithms and comparative studies using Machine Learning (ML) algorithms. Ten different most widely used algorithms, including unsupervised and supervised ML, will be employed to categorize the pollution hotspots for the Terengganu River. Besides, we also validate algorithms' accuracies by improving and changing each parameter in ML algorithms. Our results list all the accurate and efficient ML algorithms for the classification of river pollutions. These results help to facilitate river prediction using efficient and accurate algorithms in various water quality scenario.