High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Cluster Optimization for Similarity Process Using De-Duplication

Author(s):

DILEEP KUMAR KADALI , Swarnandhra College of Engineering & Technology, Seetharampuram, Narsapur-534280 ; R.N.V. JAGAN MOHAN, Swarnandhra College of Engineering & Technology, Seetharampuram, Narsapur-534280 ; M. Srinivasa Rao, Swarnandhra College of Engineering & Technology, Seetharampuram, Narsapur-534280

Keywords:

Big Data, Foliage Images, De-Identification, Support Vector Optimization and Map Reduce

Abstract

Over the decades, voluminous of data revolution where nearly every aspect of computer engineering is being driven by large-data processing and analysis. Valid data is important for accessing system without De-Identification is a well-known technique for recognizable system which provides the privacy for individuals. This system incorporates large data sets i.e., Big Data by forming clusters. In hierarchical clustering, the output is a tree giving a sequence of clustering, with each cluster being a partition of the dataset. The major drawbacks of the clusters in automatic research with the choice of the most relevant features are not compared to the similarity process (Duplication). However, the system suffers to how to reduce the computational load on leaf images. In this paper, an algorithm proposed on personnel signature based foliage images to optimize the automatic feature-subset selection based T-nary clusters and also classification of clustering is used for support vector machine with the help of Map Reduce Technique. The main focus is to concentrate on foliage images i.e., particularly for high-dimensional data sets and which are not at all relevant for a given operation. Also the low-dimensional feature subspaces are used to form the number of clusters and which are used to decide the cluster centers with the most relevant features at a faster pace.

Other Details

Paper ID: IJSRDV4I60433
Published in: Volume : 4, Issue : 6
Publication Date: 01/09/2016
Page(s): 830-832

Article Preview

Download Article