Classification, recognition and feedback in text based metacommunication
Gábor Szücs, Gábor Magyar · 2011
The goal of this paper was to recognize such marks, signs in human-computer communication, which refers to state of the partner. The paper deals with two tasks: speech style classification and emotion recognition of the speakers based on only the written text of the communication; and an answer generation task based on the recognized speech style or emotion. Our work has been focused on classification and recognition by text mining method using text preparation and classification algorithm.