Efficient Clustering Uncertain Data Using Hybrid Algorithm
The problem of clustering uncertain objects whose locations are described by probability density functions (pdfs) and expected distance calculations are costly operations. The UK-Means algorithm, which was the first algorithm to handle the uncertain data/objects uncertain data distances. In this paper propose pruning techniques that are based on Voronoi diagrams to reduce the number of expected distance calculations. To reduce number of Expected Distance (ED) calculations, introduces a partial ED evaluation method and have incorporated the method in VDBiP and by combination of this algorithm is called as Hybrid Algorithm. Voronoi diagram technique also decreases the cost of calculations of uncertain data, but proposed method of this paper prompts best results.