Machine learning (ML)is an emerging field in data science that predicts insights from a specific domain. Many tropical nations including Indiaare suffering from viral discrete diseases. One of the most threatening discrete diseasesis dengue virus. Hence it should be detected earlier for preventing the further spread around the prone area.In addition with exhausted pressure of covid-19 patients in hospitals early detection of dengue has become the real challenge to the physician due to similar symptoms of these severe viruses. Moreover currently existing algorithm failed to predict the disease due to lack of accuracy. The proposed method is to develop anefficient and accurate dengue prone area prediction classifier model.This is a multi-class classification model based on random forest algorithm which detects dengue prone area more effectively.The performance of the proposed system has been achieved using 12-folds cross-validation and split techniques. The performance metrics used in the proposed system includes precision, f-measures and accuracy. The results prove that the proposed system is more progressive by improvising in about 82.75 percent in accuracy, 85.7 percent in precision when compared with the existing system.

Random forest, Decision tree, Dengue diagnosis, Machine learning