Development of Prediction and Forecasting Model for Dengue Disease using Machine Learning Algorithms

Globally, Dengue is one of the most quickly spreading vector-borne viral sicknesses with an expanding number of territories in danger. Many researchers have worked on different measures to control and prevent the spread of disease. The main objective of the research is to develop a forecast model to control the outbreak of dengue disease that will give an opportunity for medical professionals in designing, planning and handling the disease at an early stage. Moreover, the improvement of the assortment of strategies for determining and predictive modeling utilizing measurable, numerical examination of machine learning was studied. There are mainly six issues need to be solved in determination of dengue disease, those are exploring data sources, analyzing data sources, techniques for data preparation, data representation, dengue forecasting models and evaluation approaches. A major limitation of the traditional methods is that these methods need large volumes of data for data processing, to improve the dynamic characteristics. From the review of existing methods, it can be clearly stated that the K-means clustering method with fuzzy based system has high accuracy and it significantly improves the analysis/prediction of dengue disease. The k-means clustering algorithm separates the dengue diseased patient records into k divisions. As the dengue dataset were fully clustered, k-means clustering method improves the analysis or prediction of dengue disease. Similarly, the fuzzy based system The input factors and changing over these informational factors into fuzzy membership functions will make a better decision making in predicting dengue forecasting model. Thus, the issues stated from comprehensive research provide a useful platform for public health research and epidemiology.