Evaluation of Different Boosting Algorithms in Water Quality Analysis at Laguna Lake

In this paper, the researchers compared different boosting algorithms in the water quality index classification. Weighted Arithmetic Water Quality Index (WAWQI) method was used in WQI computation. To evaluate the classification models, the researchers compared its speed and its performance such as accuracy, precision, recall and f1-score. It showed that CatBoost got the highest accuracy of.938 with a learning rate of 0.10, then XGBoost with.93. In speed test, XGBoost is the fastest with processing speed of 1.564s, followed by HistGradientBoost with 2.138s. The findings recommend that XGBoost overall is the best when balancing the performance and the speed of classification model. In addition, DAO 2016-08 and DAO 2021-19 were used for water quality classification. It helps to identify water bodies and its beneficial use. The work makes a significant addition to the management of water quality since it offers a useful system for identifying water quality indices. These results can help decision-makers in order to maintain not only the health of Laguna de Bay but also its sustainability in long years.