Network Anomaly Detection

Final Year Project FYP16021

READ MORE

INTRODUCTION :

Project

Network has brought convenience to the world by allowing fast transformation of data, but it also exposes a number of vulnerabilities. With anomaly detection systems, the outliers of packets can be detected and computers are prevented from attacks. Some anomaly detection systems found in literature are based on data mining methods. Recently, as deep learning becomes a popular area of research, we propose using deep learning as the model for anomaly detection in this project.

Internet Security

The Internet is evolving and it has revolutionized the world since the World Wide Web was invented. The usage of the Internet has become necessary in various areas. Through the Internet, we are able to gain access to remote hosts, retrieve data and operate on the hosts. This simplifies our day-to-day life, but without appropriate security measures, it is likely that the systems would be compromised, causing individuals and companies suffering from great loss. Intruders may gain unauthorized privileges, or simply overload the server to make it unavailable. Both of these may incur great loss for the system owners.

PROJECT SCHEDULE AND DELIVERABLES:

Time Content
September • Study the theory of deep learning
• Familiarize myself with deep learning model constructions
4 October 2016 Deliverables of Phase 1
(Inception)
Detailed project plan
Project web page
October - November Development for application protocol identification
December Development for anomaly detection (I)
11-15 January 2017 First Presentation
24 January 2017 Deliverables of Phase 2
(Elaboration)
Intermediate report
Presentation slides
February - Mid March Development for anomaly detection (II)
Mid March – Mid April Study the performance of anomaly detection
17 April 2017 Deliverables of Phase 3
(Construction)
Final report
18-22 April 2017 Final presentation
Presentation slides / Supplementary
2 May 2017 Project exhibition
Poster

Contact Us :

Dr. S. M. Yiu

Supervisor
Email: smyiu AT cs.hku.hk

Tien Hsuan Wu

Student
Email: thwu AT hku.hk