Open AccessA Design of Intrusion Detection using Modified Bat Algorithm and Deep Autoencoder Network
Author(s) -
Soundari D.V M.E.*,
R Kuralarasi.,
S Kalieshwari,
M Kanimozhi,
N Kanimozhi
Publication year - 2020
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.e2788.049620
Subject(s) - autoencoder , computer science , intrusion detection system , artificial intelligence , deep learning , algorithm , classifier (uml) , bat algorithm , data mining , machine learning , particle swarm optimization
In this world, Intrusion Detection is more popular for preparing the network security systems. In current trend of increasing security system, there is a demand for Intrusion Detection. With these clarifications need to find a huge Data measurement, high speed traffic’s and frequent forms of threats. In this work, Intrusion Detection is done by Deep Auto-Encoder network (DAEN) and Modified BAT algorithm (MBA). Our approach improves the Deep Auto Encoder (DAE) classifier by manipulating the benefits of an additional process encourage through the atmosphere of microbats (Bat Procedure). The core aim of this work is to select the features based on Modified Bat Algorithm. Towards examine the model, using the NSL-KDD data’s and the survey of Modified Bat Algorithm will be discussed. Moreover, these methods do well to improve DAEN classifier and get reliable performance in standing of accuracy (96.06%), attack detection rate (95.05%).