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

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