Enhancing of Red Tide Blooms Prediction using Ensemble Train

Red tide is a natural phenomenon temporary blooming harmful algal with changing sea color from normal to red, which fish and shellfish die en masse. It also give a bad influence to coastal environment and sea ecosystem. The damage of sea farming by a red tide has been occurred each year which it cost much to prevent disasters of red tide blooms. Red tide damage and prevention cost of red tide disasters can be minimized by means of prediction of red tide blooms. In this paper, we proposed the red tide blooms prediction method using ensemble train. The proposed method use the bagging and boosting ensemble train methods for enhancing red tide prediction and forecast. The experimental results demonstrate that the proposed method achieves a better red tide prediction performance than other single classifiers.