A HYBRID INTELLIGENT FORECASTING MODEL TO DETERMINE MALARIA TRANSMISSION

Abstract

Strategies for fighting infectious disease rely on vector control, transmission blocking and medical assistance. Malaria is an infectious disease. According to World Health Organization 200 to 300 million of people are being affected with malaria every year and about 3 million people face fatal death yearly (WHO, 2013). This sporadic occurrence of malaria diseases in human has pushed the need to develop computational approaches for predicting in advance the occurrence of malaria diseases. The need to forecast in advance the occurrence of malaria disease and its outbreak will be helpful to take appropriate actions by individuals, World Health Organizations and Government Agencies and its devastating impact will be reduced. This research work proposes a coupling of Support Vector Machine (SVM), with Fuzzy Logic System (FLS) to determine the rate of malaria transmission. An hybrid forecasting model will be developed by conjoining Support Vector Machine, with fuzzy logic system to form a single model called SVM-FLS. The hybridization is done in such a way that the unique features of Support Vector Machine and Fuzzy Logic System (SVM-FLS) are captured, and then the weakness of one is strengthened by the other. Monthly crosssectional surveys between January 2012 and December 2015(4 years) of malarial incidences will be collected from sample health centers in Minna Metropolis, Nigeria. Monthly averages of rainfall, temperature and relative humidity altogether with monthly malaria incidences will be considered as input variables. The proposed SVM-FLS will be compared with other existing models like Artificial Neural Networks(ANN), Fuzzy Logic System(FLS) and also with Support Vector Machine(SVM) to check its robustness and viability. Their statistical analysis will be conducted and their results will be compared.