Variable Selection Using Nearest Neighbor Rule in Discriminant Analysis of Dichotomous Data
Kyubark Shim
Kyubark Shim, Department of Bigdata and Applied Statistics, Dongguk University, College of Science and Technology, Gyeongju, Korea.
Manuscript received on 08 May 2019 | Revised Manuscript received on 19 May 2019 | Manuscript Published on 23 May 2019 | PP: 1219-1222 | Volume-7 Issue-6S5 April 2019 | Retrieval Number: F12090476S519/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In recent years, as interest in discriminant analysis of big data has increased, research on this field is active. The use of Big Data is also essential for politics and social issues that are sensitive to public opinion. Classifying precisely which respondents belong to which group is very helpful for policy formulation. Kim et al. (2013) propose an intelligent Voice of customer(VOC) analyzing system based on opinion mining to discriminant the unstructured VOC data automatically and determine the polarity as well as the type of VOC. In this paper, based on the previous studies, we selected the variables that can minimize the discrimination error by using the nearest neighbor method to the discriminant analysis of the formalized data.
Keywords: Big Data, Dichotomous Data, Discriminant Analysis, Nearest Neighbor, Variable Selection.
Scope of the Article: Data Analytics