Home // ICAS 2019, The Fifteenth International Conference on Autonomic and Autonomous Systems // View article

Design of Autonomous Systems for Cybersecurity Threat Detection Using Deep Learning

Authors:
Strahil Sokolov

Keywords: cybersecurity; autonomous threat detection; deep learning

Abstract:
In this paper, an approach is proposed for designing autonomous systems featuring machine learning and neural networks for cybersecurity threat detection. It is proposed that neural models are trained on monitoring data obtained from cloud environments that service enterprise applications. Cybersecurity is a hot topic and a broad field of science that spreads over activities, such as protecting infrastructure, computers and servers, industrial and telecommunications equipment, applications and data. All modern networks are capable of substantial throughput due to enormous volumes of generated traffic. A design is proposed for autonomous threat detection systems, which are based on combining traditional and deep neural networks for cloud monitoring data analysis and an algorithm for combining classifier results. The proposed autonomous system design delivers promising results, that are comparable to existing approaches and can become useful in enterprise cloud applications.

Pages: 46 to 50

Copyright: Copyright (c) IARIA, 2019

Publication date: June 2, 2019

Published in: conference

ISSN: 2308-3913

ISBN: 978-1-61208-712-2

Location: Athens, Greece

Dates: from June 2, 2019 to June 6, 2019