Abstract

The traditional dengue surveillance and dengue control hindered by its lack of timely responsiveness and usage of obsolete data. Over the years, various dengue prediction models have been developed using machine learning techniques. This technique enables the computers to learn without being explicitly programmed. The main objective of this paper is to explore the opportunities and benefits of using this technique to provide forecast estimates of Aedes mosquito abundance on a smaller scale using data collected at the target area. The procedure in development of mosquito predictor models using SVM and MLR algorithms are field data collection at the target area, data preparation, features or independent variables selection, building and training of models and model performance.