Linguistic Profiling and Behavioral Drift in Chat Bots

23rd Midwest Artificial Intelligence and Cognitive Science Conference 2012

Nawaf Ali ; Derek Schaeffer ; Roman V. Yampolskiy

Others  :  http://ceur-ws.org/Vol-841/submission_22.pdf
PID  :  43862

来源: CEUR

【 摘 要 】

When trying to identify the author of a book, a paper, or aletter, the object is to detect a style that distinguishes oneauthor from another.With recent developments inartificial intelligence, chat bots sometimes play the role ofthe text authors.The focus of this study is to investigatethe change in chat bot linguistic style over time and itseffect on authorship attribution.The study shows that chatbots did show a behavioral drift in their style.Resultsfrom this study imply that any non-zero change in lingualstyle results in difficulty for our chat bot identification process.