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

Dimension reduction of Data instance: An inspection

Author(s):

Ujjawal Singh , SSTC(SSGI),BHILAI; Megha Mishra, SSTC(SSGI),BHILAI

Keywords:

Supervised learning, Unsupervised learning, Feature selection

Abstract

Data mining application has massive advantages, as historical data have huge number of features. Feature selection is an essential role in improving the eminence of learning algorithms in data mining and machine. This has been broadly deliberated in supervised learning, whereas it is still comparatively infrequent researched in case of unsupervised learning. Every data mining application has common issue; dataset has huge number of features which is immaterial or redundant to the data mining job in hand which pessimistically affects the performance of the fundamental learning algorithms, and makes them less efficient. Henceforth reducing the dimensionality of dataset is primary and important job for data mining applications and machine learning algorithms so that computational burden of the learning algorithms can be minimized. In this paper we will discuss different feature selection algorithms so as to find out factors which affect the performance of existing algorithm so that we can move further for researching another novel method for data mining application.

Other Details

Paper ID: SPDM036
Published in: Volume : 1, Issue : 2
Publication Date: 01/11/2015
Page(s): 25-28

Article Preview




Download Article

Share