Conversational agents emulate human communication in order to provide a more natural user interface to applications or simply to begin a dialog with a user. Prior studies of this software have focused upon individual systems, but few have provided comparisons of their capabilities. Here, we evaluate three such programs (Tutor Mike, Cleverbot, and Jabberwacky). Results show that there was a significant difference among the three chatbots in terms of perceived friendliness and knowledge about the topic under discussion, but users did not believe the systems were very humanlike and probably would not use them again.
Download PDF:
Journal Area:
None