Comparative Study of Clustering Algorithms by Conducting a District Level Analysis of Malnutrition

摘要

Bihar is a state that faces the highest level of malnutrition in India, leading to a lot of serious health problems. This study was taken up to better understand the malnutrition factors in Bihar by assessing and comparing results of two clustering algorithms. Unstructured data was obtained from Government reports, pertaining to malnutrition indices, demographic, social and medical factors that cause malnutrition. In the first stage the variables from non-clustered data was correlated with malnutrition indices. Further on, the data was broken into clusters. This was carried out by the Rapid Miner Studio using k-means and Hierarchical agglomerative clustering. Subsequently, each cluster was again analyzed using the software and the correlation results were compared. Significant variation was observed in most of the correlations in the data sets obtained by executing the two algorithms. Choosing a more suitable algorithm that best represents malnutrition in Bihar would thus be very useful. This has significant implications in policy making for malnutrition control, through identifying the most relevant variables/factors responsible.