Dengue is one of the most common and widespread infectious illnesses in humans transmitted by female Aedes albopictis. The prevalence of Dengue cases has increased substantially leading to human morbidity. Inadequate availability of healthcare professionals and inaccessibility to healthcare institutions have aggravated the problem. The traditional medical technologies are too antiquated to serve the purpose. The innovative latest technologies like Internet of Things (IoT), Cloud Computing, Fog Computing have made real-time and remote healthcare possible with huge success. In this paper, an IoT based Fog–Cloud enabled system for monitoring, assessment and control of Dengue Fever has been proposed. IoT sensors acquire data about a large spectrum of health as well as environmental factors that contribute to infection. The battery constrained sensors set their sampling rate according to the degree of cruciality that saves power to make battery long lasting. The Fog layer employs Support Vector Machine (SVM) for Dengue infection evaluation with least latency and sends alerts including precautionary measures to the users, hospital officials and government agencies. Moreover, the proposed system utilizes Temporal Network Analysis (TNA) and Google map service to categorize areas as infected, uninfected or risk prone. The experimental results are evaluated by a number of analytical parameters to investigate the effect of proposed system. SVM performs the best in terms of accuracy, recall, specificity, precision and f-measure with values 93%, 95%, 89%, 94% and 95% respectively. Furthermore, TNA based outbreak assessment gives valuable inputs for the government institutions to control the outbreak.

登革热是由女性白纹伊蚊传播的人类最常见和最广泛的传染病之一。登革热病例的患病率大大增加，导致人类发病。医护人员的不足和医护机构的无所作为加剧了这个问题。传统医疗技术过时了，无法达到目的。诸如物联网（IoT），云计算，雾计算之类的创新性最新技术已使实时和远程医疗保健取得了巨大成功。在本文中，提出了一种基于物联网的基于Fog-Cloud的系统，用于监控，评估和控制登革热。物联网传感器可获取有关大量健康以及导致感染的环境因素的数据。电池受限的传感器根据关键程度设置采样率，以节省电量，从而使电池持久耐用。雾层采用支持向量机（SVM）进行登革热感染评估，从而减少了等待时间，并向用户，医院官员和政府机构发送了包括预防措施在内的警报。此外，所提出的系统利用时间网络分析（TNA）和Google地图服务将区域划分为受感染，未感染或易发风险的区域。通过许多分析参数评估实验结果，以研究所提出系统的效果。SVM在准确性，召回率，特异性，准确性和f量度方面表现最佳，分别为93％，95％，89％，94％和95％。此外，