Creating dictionary for the field of human-robot interaction

The paper is about the sentiment analysis within human expressions mainly in the form of texts. The sentiment analysis focuses on two kinds of sentiment: an opinion and emotion. The purpose is to provide the discovered opinions and emotions to a robot. Consequently, the robot will be able to adapt its behavior to the actual emotional situation. The results of the research can be utilized in work with children, mainly those suffered from a neurodevelopmental disorder such as autism. The paper introduces basic approaches to the emotion discovering. The design of the original approach to emotion classification based on the dictionary is presented and tested on Aesop fables.