Call For Paper Volume:4 Issue:10 Oct'2017 |

Distribution Similarity based Data Partition and Nearest Neighbor Search on Uncertain Data

Publication Date : 31/12/2014



Author(s) :

D. Parthipan , Dr. M. Moorthi Ph.D.


Volume/Issue :
Volume 1
,
Issue 6
(12 - 2014)



Abstract :

Databases are build with the fixed number of fields and records. Uncertain database contains a different number of fields and records. Clustering techniques are used to group up the relevant records based on the similarity values. The similarity measures are designed to estimate the relationship between the transactions with fixed attributes. The uncertain data similarity is estimated using similarity measures with some modifications. Clustering on uncertain data is one of the essential tasks in mining uncertain data. The existing methods extend traditional partitioning clustering methods like k-means and density-based clustering methods like DBSCAN to uncertain data. Such methods cannot handle uncertain objects. Probability distributions are essential characteristics of uncertain objects have not been considered in measuring similarity between uncertain objects.  The customer purchase transaction data is analyzed using uncertain data clustering scheme. The density based clustering mechanism is used for the uncertain data clustering process. This model produces results with minimum accuracy levels. The clustering technique is improved with distribution based similarity model for uncertain data. The nearest neighbor search technique is applied on the distribution based data environment. The system is designed using java as a front end and oracle as a back end.


No. of Downloads :

23


Indexing

Web Design MymensinghPremium WordPress ThemesWeb Development

Distribution Similarity based Data Partition and Nearest Neighbor Search on Uncertain Data

December 19, 2014