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

Feature Selection for Opinion Mining using Binary Cuckoo Search Algorithm

Author(s):

S. M. Hemalatha , KONGU ENGINEERING COLLEGE; C. S. Kanimozhi Selvi, KONGU ENGINEERING COLLEGE

Keywords:

Feature Selection, Opinion Mining, Binary Cuckoo Search Algorithm

Abstract

Feature selection is a process of identifying the related subset with less intricacy. It is used to handle the optimization problem in classification. The dataset may comprise enormous number of features so it is difficult to identify the related features because the dataset may contain unrelated features. To overcome this problem, a binary cuckoo search based feature selection algorithm (BCSA) is proposed. It enhances the process of feature selection and produces the best optimal feature subset which increases the predictive accuracy of the classifier with minimum number of features. It improves the search space both global and local. In this BCSA is used as a feature selector and yields the feature subset and SVM classifier is used to estimate the feature subset produced. The analysis result show that the SVM classification yield enhanced accuracy when binary cuckoo search algorithm is used for feature selection.

Other Details

Paper ID: IJSRDV6I20436
Published in: Volume : 6, Issue : 2
Publication Date: 01/05/2018
Page(s): 976-981

Article Preview

Download Article