Academic
Publications
Methods for chatbot detection in distributed text-based communications

Methods for chatbot detection in distributed text-based communications,10.1109/CTS.2010.5478478,John P. McIntire,Lindsey K. McIntire,Paul R. Havig

Methods for chatbot detection in distributed text-based communications   (Citations: 1)
BibTex | RIS | RefWorks Download
Distributed text-based communications (e.g., chat, instant-messaging) are facing the growing problem of malicious “chatbots” or “chatterbots” (automated communication programs posing as humans) attempting social engineering, gathering intelligence, mounting phishing attacks, spreading malware and spam, and threatening the usability and security of collaborative communication platforms. We provide supporting evidence for the suggestion that gross communication and behavioral patterns (e.g., message size, inter-message delays) can be used to passively distinguish between humans and chatbots. Further, we discuss several potential interrogation strategies for users and chat room administrators who may need to actively distinguish between a human and a chatbot, quickly and reliably, during distributed communication sessions. Interestingly, these issues are in many ways analogous to the identification problem faced by interrogators in a Turing Test, and the proposed methods and strategies might find application to and inspiration from this topic as well.
Published in 2010.
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...We attempted to study such questions in a previous work [12], using a different (and considerably smaller) dataset...
    • ...As in our previous work [12], we gathered our chat data from the publicly-available transcripts of the Loebner Prize in...
    • ... are suspicious as to the other communicators’ identities from the start, and many of the conversations might be rightly classified as “interrogations” as opposed to merely “normal, everyday conversations.” Nonetheless, since both the human participants (confederates) and bots faced similar interrogations by the human judges, we do not believe this to be a fatal flaw to our analyses and interpretations; indeed, in our previous work [12], ...
    • ...Although in our previous work [12] we used data only from the 2008 competition, in this work we analyzed five separate competitions, from the years 1996, 1997, 2004, 2005, and again from 2008...
    • ...These results directly support our previous work in [12], the results of Gianvecchio...
    • ...This result is in contrast to our previous findings (with a smaller sample) that suggested message sizes to possibly be a distinguishing metric [12]...

    John McIntireet al. Graphical and statistical communication patterns of automated conversa...

Sort by: