Publication

Abstract : In the structure of nature, we believe that there is an underlying knowledge in all the phenomena we wish to understand. Mainly in the area of epidemiology we often tend to seek the structure of the data obtained, pattern of the disease, nature or cause of its emergence among living organisms. Sometimes, we could see the outbreak of disease is ambiguous and the exact cause of the disease is unknown. A significant number of algorithms and methods are available for clustering disease data. We could see that literature has no traces of including indeterminacy or vagueness in data which has to be much concentrated in epidemiological field. This study analyzes the attack of dengue in 26 districts of Sri Lanka for the period of seven years from 2012 to 2018. Clusters with low risk, medium risk and high risk areas affected by dengue are identified. In this paper, we propose a new algorithm called Neutrosophic-Fuzzy Hierarchical Clustering algorithm (NFHC) that includes indeterminacy. Proposed algorithm is compared with fuzzy hierarchical clustering algorithm and hierarchical clustering algorithm. Finally the results are evaluated with the benchmarking indexes and the performance of the clustering algorithm is studied. NFHC has performed a way better than the other two algorithms.