An Efficient Approach for Clustering Uncertain Data Mining Based on Hash Indexing and Voronoi Clustering
Authors:
- Samir AjaniSASamir Ajani
- Mangesh WanjariMWMangesh Wanjari
- 1 September 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 486-490
Abstract
Recently the classifying uncertain data in spatial database used in data mining attract a main attention by researchers. The main task is to handle the uncertainty of the data in order to classify or cluster it. In order to classify or cluster the valid or certain data, there are various techniques like DTL, Rule based Classification, Naive Bayes Classification and many more techniques. It's easy to classify the certain data but classification of the uncertain data is bit difficult. Generally K-means algorithm is used to make clusters of uncertain of uncertain data but it increases overhead and computation time. Hence to improve the performance of K-means algorithm we are proposing a technique in which K-means is combined with Voronoi clustering but this will increase the prune overhead so to reduce this prune overhead we will add Hash indexing on th uncertain data object. Our technique of combining K-means with hash indexing and Vornoi diagram results better than older technique used for clustering based on K-Means algorithm.Keywords
This publication has 4 references indexed in Scilit:
- Automatic classification of uncertain data by soft classifier2011
- UV-diagram: A Voronoi diagram for uncertain data2010
- A Survey of Uncertain Data Algorithms and ApplicationsIEEE Transactions on Knowledge and Data Engineering, 2008
- An Efficient Distance Calculation Method for Uncertain Objects2007