Speaker name: Tom Szyszko
Talk’s title and abstract:
Automated Victim: Final Year Projects at Internet Scale
Netcraft operates an industry-leading feed of validated phishing websites, which is licensed by all leading web browsers to protect their users. Over 56 million unique phishing sites have been detected and blocked by Netcraft to date. The scale of the problem is vast: every day Netcraft processes millions of suspect URLs in order to identify new threats. URLs found to be phishing are included in the feed, whilst harmless URLs are discarded. Prompt classification of each site is essential in order to protect users from new attacks as soon as possible. To this end, Netcraft automates its classification of suspected phishing sites to the fullest extent. The "Automated Victim" approach to classification began as a DoC final year project, and now runs in production at Netcraft. It automatically mimics the behaviour of a human interacting with a phishing site - filling form fields, scrolling, and uploading files - in order to perform an accurate classification. We demonstrate the effectiveness of this approach, whilst outlining the challenges of scaling up from an undergraduate project to a distributed production system.
Bio: Tom Szyszko is a Computer Scientist at Netcraft, where he works on the development of automated phishing site classification techniques. He graduated from Imperial College London in 2018 with an MEng in Computing.