Modified Artificial immune system as Feature Selection

  • Jamal H. Assi Computer Science Department, Technology University, Baghdad, Iraq.
  • Ahmed T. Sadiq Computer Science Department, Technology University, Baghdad, Iraq.

Abstract

Feature selection algorithms play a big role in machine learning applications. There are several feature selection strategies based on metaheuristic algorithms. In this paper a feature selection strategy based on Modified Artificial Immune System (MAIS) has been proposed. The proposed algorithm exploits the advantages of Artificial Immune System AIS to increase the performance and randomization of features. The experimental results based on NSL-KDD dataset, have showed increasing in performance of accuracy compared with other feature selection algorithms (best first search, correlation and information gain).

Published
Apr 29, 2018
How to Cite
ASSI, Jamal H.; SADIQ, Ahmed T.. Modified Artificial immune system as Feature Selection. Iraqi Journal of Science, [S.l.], p. 733-738, apr. 2018. ISSN 2312-1637. Available at: <http://scbaghdad.edu.iq/eijs/index.php/eijs/article/view/250>. Date accessed: 26 may 2018.
Section
Computer Science