BayeShield: A Conversational Anti-Phishing User Interface

In this paper we present BayeShield, a novel anti-phishing tool that uses a conversational approach to partner with users in determining whether a website is phishing when the website is suspicious but not blacklisted. We describe the iterative user-centered development of BayeShield's user interface, discussing its evolution and the design principles we followed. In an empirical evaluation, BayeShield performed better than Firefox 2.0 in preventing participants from entering information on phishing sites. In a second portion of the study, we evaluate BayeShield's usability and obtain positive results including high user satisfaction ratings, and a high-level of engagement as demonstrated by perceived duration of tasks being lower than actual durations. In addition, we learned user characteristics that affect the likelihood users will enter information on phishing websites. Author