Clustering with density based initialization and Bhattacharyya based merging

Centroid based clustering approaches, such as k-means, are relatively fast but inaccurate for arbitrary shape clusters. Fuzzy c-means with Mahalanobis distance can accurately identify clusters if data set be modelled by a mixture of Gaussian distributions. However, they require number apriori and bad initialization cause poor results. Density methods, DBSCAN, overcome these disadvantages. may perform poorly when the dataset is imbalanced. This paper proposes method, named density Bhattacharyya merging on fuzzy clustering. The carried out estimation adaptive bandwidth using k-Nearest Orthant-Neighbor algorithm to avoid effects imbalanced local peaks point clouds constructed used initial cluster centers We use measure Jensen inequality find overlapped Gaussians merge them form single cluster. experiments variety datasets show that proposed has remarkable advantages especially arbitrarily shaped sets.