Open AccessCNN Intrusion Detection for Threat Analysis of a Network
Author(s) -
Tressa Michael
Publication year - 2021
Publication title -
türk bilgisayar ve matematik eğitimi dergisi
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.218
H-Index - 3
ISSN - 1309-4653
DOI - 10.17762/turcomat.v12i3.1683
Subject(s) - computer science , computer security , internet of things , intrusion detection system , intrusion , convolutional neural network , artificial intelligence , data science , risk analysis (engineering) , business , geochemistry , geology
The technological advancement realized in the discovery and embrace of both IoT and IIoT is totally indispensable. Many systems and subsystems both robust and miniaturized have made their existence into the technical arena due to IoT. It goes without saying that IoT has brought into light very diverse benefits that cut across universal applications.However, the pre-requisite of a network channel existence for an IoT operation to be successful is the only pitfall that this essentially unique system possesses. There is a significant amount of danger associated with transmission networks. They have very substantial susceptibility to both online and offline threats by malicious cyber intentions.This paper focuses on the analyses of the threats posed to these IoT networks through Artificial Neural Networks. Specifically, a model is trained through recurrent and convolutional neural network to do intensive analysis on the threat intensity, type and threat source for data logging purposes. The Intruder detection system (IDS) explored in this paper registers a success rate of 99% based on the empirical data posed to the model.