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

A Review on FIUT Based Parallel Mining in Mapreduce

Author(s):

Parvathi S J , GSSSIETW; Asha Rani M, GSSSIETW

Keywords:

Frequent Itemsets, HaDoop Cluster, Load Balance, Map Reduce, Data Mining

Abstract

Existing parallel mining calculations for frequent itemsets do not have a component that enables programmed parallelization, stack adjusting, information circulation, and adaptation to internal failure on large groups. As an answer for this issue, we plan a standard allel visit itemsets mining calculation called FiDoop using the MapReduce programming model. To accomplish packed capacity and abstain from building contingent example bases, FiDoop consolidates the successive things ultrametric tree, rather than conventional FP trees. In FiDoop, three MapReduce occupations are actualized to finish the mining errand. In the essential third MapReduce work, the mappers autonomously disintegrate itemsets, the reducers perform blend operations by constructing small ultrametric trees, and the real mining of these trees separately. We execute FiDoop on our in-house Hadoop cluster. We demonstrate that FiDoop on the bunch is touchy to information distribution and measurements, in light of the fact that itemsets with various lengths have diverse disintegration and development costs. To improve FiDoop's execution, we build up a workload adjust metric to quantify stack adjust over the bunch's figuring hubs. We create FiDoop-HD, an augmentation of FiDoop, to accelerate the mining execution for high-dimensional information examination. Extensive experiments utilizing genuine heavenly ghostly information demonstrate that our proposed arrangement is capable and adaptable.

Other Details

Paper ID: IJSRDV4I90132
Published in: Volume : 4, Issue : 9
Publication Date: 01/12/2016
Page(s): 157-159

Article Preview

Download Article