Construction of the EEG Emotion Judgment System Using Concept Base of EEG Features

摘要

For a robot converse naturally with a human, it must be able to accurately gauge the emotional state of the person. Techniques for estimating emotions of a person from facial expressions, intonation and speech content have been proposed. This paper presents a technique for judging the emotion of a person using EEGs. The system of judging the emotion from EEGs is called EEG Emotion Judgment System, and constructed Concept Base for reducing noise in this paper. Accuracy of emotion judgment using EEG features of all subjects was 57.6% and using leave-one-out cross validation was 30.8%. Although performance accuracy remains low, continued development is required through further development of methods for both reducing different variety of noise mixed in with EEGs.