Study of K-Means Clustering Algorithm for Identification of Dengue Fever Hotspots

摘要

In India, several dengue outbreaks were reported like 2015 dengue outbreaks, increasing the need for control and prevention of such outbreaks. Cluster analysis is one of the most extensively used statistical data analysis due to its extensive use in machine learning. Clustering techniques can be integrated to determine potential real-world areas of infectious diseases. This paper presents the application of K-means in disease surveillance. K-means is one of the least difficult unsupervised learning calculations. It groups the dataset through a specific number of clusters. The primary thought is characterized k-centroids, one for each cluster. In India, several dengue outbreaks were reported like 2015 dengue outbreaks, thereby increasing the need for control and prevention of such outbreaks. This paper presents an experimental evaluation of the effectiveness of K-means using data of reported Dengue Fever (DF) which was maintained in the Health Department of Municipal Corporation of Delhi (MCD). A total of 4713 cases were seen in years 2011, 2012 and 2013. The paper successfully detected hotspots of DF and calculated the Silhouette coefficients to validate the clusters.