MailTrout: a machine learning browser extension for detecting phishing emails

Paul Boyle, Lynsay A. Shepherd*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1869 Downloads (Pure)


The onset of the COVID-19 pandemic has given rise to an increase in cyberattacks and cybercrime, particularly with respect to phishing attempts. Cybercrime associated with phishing emails can significantly impact victims, who may be subjected to monetary loss and identity theft. Existing anti-phishing tools do not always catch all phishing emails, leaving the user to decide the legitimacy of an email. The ability of machine learning technology to identify reoccurring patterns yet cope with overall changes complements the nature of anti-phishing techniques, as phishing attacks may vary in wording but often follow similar patterns. This paper presents a browser extension called MailTrout, which incorporates machine learning within a usable security tool to assist users in detecting phishing emails. MailTrout demonstrated high levels of accuracy when detecting phishing emails and high levels of usability for end-users.
Original languageEnglish
Title of host publicationProceedings of the 33rd British Human Computer Interaction Conference
Subtitle of host publicationPost-Pandemic HCI – Living digitally
PublisherAssociation for Computing Machinery (ACM)
Number of pages12
Publication statusAccepted/In press - 20 Jun 2021
Event33rd British Human Computer Interaction Conference: Post-Pandemic HCI – Living digitally - University of West London, London, United Kingdom
Duration: 19 Jul 202121 Jul 2021
Conference number: 33rd


Conference33rd British Human Computer Interaction Conference
Country/TerritoryUnited Kingdom
Internet address

Cite this