Predicting Coral Reef Bleaching through Machine Learning

Coral reef bleaching is a growing, ongoing issue that has yet to be tackled completely. One way to prevent these events is to make data-driven decisions, which can be done through data science. This paper uses machine learning to potentially predict future coral bleaching so that actions can be taken before the event’s occurrence. Three machine learning algorithms are used, namely Support Vector Machines (SVM), Decision Trees, and Naïve Bayes. In this paper, we use both numerical and categorical datasets from BCO-DMO. The performances of the models, such as SVM, Decision Tree, and Naïve Bayes are evaluated to reveal which algorithm performed best in this application. The findings reveal that SVM had the best overall performance in predicting coral bleaching events, with an accuracy score of 93 percent for the BCO-DMO dataset. Further research can be done to expand and improve the reliability of these results.