Classification of infectious diseases via hybrid k-means clustering technique

Abstract

Identifying groups of objects that are similar to each other but different from individuals in other groups can be intellectually satisfying, profitable, or sometimes both. Kmeans clustering is one of the well known partitioning algorithms. But basic K-means method is insufficient to extract meaningful information and its output is very conscious to initial positions of cluster centers. In this paper, data of infectious diseases were analyzed with the hybrid K-means clustering technique. This method is developed to preprocess the dataset that will be used in the K-means clustering problems. Specifically, it performs K-means clustering on preprocessed dataset instead of raw dataset to remove the impact of irrelevant features and selection of good initial centers. The experimental results revealed that all the three water related diseases are grouped together in one cluster for both KGHK and FMCK data sets. They also show the high prevalence compared to airborne particle related diseases in the other group. The study concludes that K-means clustering method provides a suitable tool for assessing the level of infectious diseases.