Visualization of dengue incidences for vulnerability using K-means

Dengue is the world´s most rapidly spreading and geographically widespread arthropod-borne disease. Dengue epidemics are observed to be larger, more frequent and associated with more severe disease than they were in the past. To control the incidence of the disease, it is important to be able to identify the hot-spots localized regions of high incidences. This work focuses on identifying hot-spots of dengue using the K-means clustering algorithm. Data is collected from the state of Selangor in Malaysia from 2013 to 2014. Visualization of dengue vulnerability is obtained via Gaussian mixture models fitted using K-means algorithm. Results demonstrate the ability to render visualization for the vulnerability of dengue incidences on the basis of high density and low density cluster using Gaussian mixture and K-means algorithm.