PROBABILISTIC INFORMATION BASED CLUSTERING INVESTIGATION OF PROBABILITY DISTRIBUTION SIMILARITY

COMPUTER SCIENCE ENGINEERING

Clustering is an essential task in data mining. The fundamental purpose of clustering is gathering the same object data in a massive dataset and identifying resemblances between the objects. Clustering of uncertain data is a more complicated task in both designing the similarity of data objects and implementing the efficient computational methods. Clustering uncertain data problems has been explained by utilizing many modern data mining techniques and numerous methods. Techniques have newly been convenient for clustering uncertain data based upon the conventional dividing clustering methods like k- means and density-based clustering methods like DBSCAN for uncertain data, they will be resolved by geometric distances between objects. Computing the resemblance between the data objects will be based upon a correlation distance measure and further clustered with occurrence based clustering or hierarchical clustering methods. Such methods cannot handle uncertain elements that are geometrically no conflict. In the recommended system, we could use probability that are the fundamental attributes of uncertain objects and are analyzed in measuring likeness between uncertain objects. The extremely suitable technique Kullback-Leibler divergence is employed to operations the distribution relationship between two uncertain data items. First the probability division method for a model, uncertain data object then thereafter estimate the similarity between data objects using distance metrics.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.