Value Decomposition and Dimension Selection in Multi-Dimensional Datasets using Map-Reduce Operation

Abstract

 The datasets which are in the form of object-attribute-time format is referred to as three-dimensional (3D) data sets. Clustering these three-dimensional (3D) data sets is a difficult task. So the subspace clustering method is applied to cluster the three-dimensional (3D) data sets. But finding the subspaces in the these three-dimensional (3D) dataset which is changing over time is really a difficult task. Sometimes this subspace clustering on threedimensional (3D) data sets may produce the large number of arbitrary and spurious clusters. So to cluster these three-dimensional (3D) data sets a new centroid based concept is introduced called CATS. This CATS allows the users to select the preferred objects as centroids. This algorithm is not the parallel one. So it increases the time and space requirements which are needed to cluster the three-dimensional (3D) data sets. And in CATS no optimal centroids have been chosen to cluster the three-dimensional (3D) datasets. Since the CATS clusters the data based on the fixed centroids, the CATS cannot produce the good quality clusters. So for the first time in the proposed method the CPSO technique is introduced on the three-dimensional (3D) data sets to overcome all these drawbacks which clusters the three-dimensional (3D) datasets based on the optimal centroids and also it acts as the parallelization technique to tackle the space and time complexities.

Authors and Affiliations

Preethi V*

Keywords

Related Articles

 TLS: Improving Security for Ad-Hoc Networks using Three Level Security Mechanism

 A group of large autonomous wireless nodes are connected and communicating in a P2P method on a Heterogeneous environment without defined infrastructure is named Mobile Ad-hoc network. Various mechanisms and tech...

 STUDY OF VEHICULAR AIR POLLUTION – A CASE STUDY OF DELHI

 This research is focus on how the concentration of five pollutant i.e. NO2, NO, 03, PM10, PM2.5 changes in the absence and presence of vehicle during the 24 hour a day and diurnal variation taking R K Puram locati...

Analysis and Implementation of Lifting Scheme for Image Compression

This paper proposes an improved version of lifting based 2D Discrete Wavelet Transform (DWT) VLSI architecture. In this paper, high-efficient lifting-based architecture for the 5/3 discrete wavelet transform (DWT) is p...

 A COMPACT G-SHAPED DUAL-BAND ANTENNA FOR WLAN/WI-MAX AND RFID APPLICATIONS

 In this paper a G-shaped dual-band monopole antenna with a shorted strip fed by a coupling microstrip line for wireless communication in the wireless local-area network (WLAN) band is studied. The proposed antenna...

Physical Properties of Sugarcane Pertaining to the Design of a Whole Stalk Sugarcane Harvester

Sugarcane crop plays a vital role in nation’s economy, being one of the most commercialised crops in India. The production cost of sugarcane is increasing year after year which reduces the profit margin of sugarcane gro...

Download PDF file
  • EP ID EP163898
  • DOI -
  • Views 100
  • Downloads 0

How To Cite

Preethi V* (30).  Value Decomposition and Dimension Selection in Multi-Dimensional Datasets using Map-Reduce Operation. International Journal of Engineering Sciences & Research Technology, 3(4), 1901-1907. https://www.europub.co.uk/articles/-A-163898