Cyber Meets Machine Learning
Machine learning can provide systems the ability to independently learn and adapt by extracting patterns from data without human intervention. Studies of its application in network-based intrusion detection produce favourable results when using well understood datasets. However, when applied to real-world network data, similar accuracy rates are often not achievable. This is due to several factors, including feature selection and data pre-processing. This presentation will expand on these factors, discussing their importance and application for detecting network-borne security threats.
Speakers:
Associate Professor Mike Johnstone, Edith Cowan University
Mike is an Associate Professor at the School of Science at Edith Cowan University where he teaches secure programming and advanced software engineering. As a senior member of the Security Research Institute at ECU his work on resilient systems covers secure development methodologies, wireless sensor networks and the security of IoT devices with a focus on critical infrastructure. With over 30 years of experience in ICT, he provides consultancy services in cyber security for private industry, government and research organisations and has held various IT roles including programmer, systems analyst, project manager and network manager before moving to academia.
Mike serves on various cyber-related conference committees and is the current chair of the Australian Information Security Management conference. He is also the theme lead for Network Forensics and Response to Emerging Threats in the Industry-driven, Federally-funded Cyber Security Co-operative Research Centre.
Matt Peacock, Sapien Cyber
Matthew is a machine learning engineer at Sapien Cyber, where he builds machine learning solutions to detect network-borne threats in Operational Technology Systems. Recently, Matthew has completed his PhD at the Edith Cowan University Security Research Institute, where he applied machine learning techniques for intrusion detection in Building Automation Systems using the BACnet protocol.
Participants will have the opportunity to ask questions of the speakers at the end of the presentation.
Light refreshments will be served after the presentation.
Register:
AISA members: Please sign in to register for this event
Non AISA Members: If you would like to become an AISA member you can join here
For more information about other AISA events, please visit our website www.aisa.org.au
Please note: that the event time might be displaying AEST. The event starts at 5.30pm GMT+8 and finishes at 7.30pm GMT+8.
For any queries on this event, please contact the AISA Event & Sponsorship Manager, [email protected]
Many thanks to our sponsors:
Venue Sponsor |
|
Event Sponsors |
|
|
|
|
|
|
|
|
|
|
|
|
|