Open AccessAn Efficient Approach to Microarray Data Classification using Elastic Net Feature Selection, SVM and RF
Author(s) -
C. S. N. Koushik,
A. V. Shreyas Madhav,
Rabesh Kumar Singh
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1911/1/012010
Subject(s) - feature selection , support vector machine , computer science , random forest , elastic net regularization , data mining , classifier (uml) , gene selection , microarray analysis techniques , artificial intelligence , selection (genetic algorithm) , feature (linguistics) , dna microarray , pattern recognition (psychology) , machine learning , gene , gene expression , chemistry , biochemistry , linguistics , philosophy
DNA Microarray technology forms an integral part of the bioinformatics world and provides researchers with ability to monitor a large amount of gene expressions simultaneously. The analysis of this data proves extremely beneficial in the detection of several diseases. Classification of the gene expression data obtained from microarrays is an imperative step in providing further information on the types of possible diseases present. Several classification models for microarray data have been proposed, yielding considerable results in terms of accuracy and execution time. This paper demonstrates the implementation of a microarray data classification system using Elastic Net for feature selection on two classification mechanisms. The classification of the data has been carried out using SVM and Random Forest Classifier methods. The implemented system using Elastic Net and SVM exhibits a much better performance in terms of accuracy and execution time than most of the existing systems.