Textual Emotion Recognition in chatbots
Till date no computer program has passed the Turing test although even simple conversational programs, socalled “Chatbots”, could make people believe that they are talking to another human being. The main objective of this paper is to identify some shortcomings of existing conversational agents (e.g. Chatbots) and describe our approach to model human-like conversational agent that overcomes those limitations to some extent. Our primary focus is to sense affective information from input sentence(s) by applying a cognitive theory of emotions known as the OCC model and generate both proactive and reactive responses according to the input. We thus aim at developing an emotionally intelligent computer program that not only “understands” what affective information is conveyed in textual messages, but also may provide automatic empathic response.