Downloads: 116 | Views: 182

M.Tech / M.E / PhD Thesis | Computer Science & Engineering | India | Volume 3 Issue 3, March 2014

Clustering Algorithm Based on Efficient Centroids for Schema

Beninda Churchil | Naganathan E. R | John Aravindhar.D

Abstract: Cluster analysis is an important process of grouping similar and dissimilar items. Clustering can be done efficiently by the help of finding centroid. Centroids are playing major role in efficient clustering. Clustering can be done without any outliers by grouping them efficiently based on the centroids. Clustering will be efficient for all kind of inputs. This is the main aim of our paper. We give the input in the form of two tables. We can check how efficient is the clustering if the input is in the form of tables. We use a algorithm named as Lloyd`s algorithm for making the clustering efficient. We need to find the average mean distance which should be minimum in all the cases. For this purpose we use a distance metric named as Earth mover distance metric which finds the mean value for all the entries in the table we give. Finally our clustering is analyzed with the performance check in the terms of precision and recall.

Keywords: clustering, Lloyd`s algorithm, Earthmover distance, centroids

Edition: Volume 3 Issue 3, March 2014,

Pages: 677 - 681