BROWSE
Detail
Conference Paper
자율신경계반응 지표를 이용한 세 가지 정서 분류
Cited - time in 
Share 
Share
- Authors
- Eun-Hye Jang, Yeongji Eum, Sang-Hyeob Kim, Jin-Hun Sohn
- Issue Date
- 2011-05
- Citation
- 대한인간공학회 학술 대회 (춘계) 2011, pp.359-363
- Publisher
- 대한인간공학회
- Language
- Korean
- Type
- Conference Paper
- Abstract
- Objective: The purpose of this study is to identify optimal algorithm for emotion recognition which classify three different emotional states (happiness, neutral, and surprise) using physiological features. Background: Recent emotion recognition studies have tried to detect human emotion by using physiological signals. It is important for emotion reconition to apply on human-computer interaction system for emotion detection. Method: 217 students participated in this experiment. During three different emotional stimuli are presented to participants, ANS responses(EDA, SKT, ECG, Respiration, and PPG) as physiological signals were measured for 1 minute as baseline and for 1-1.5 minutes during emotional state. The obtained signals were analyzed for 30 seconds from the baseline and the emotional state. Participants assessed the induced emotion on emotional assessment scale after emotional stimuli presentation. Analysis for emotion classification were done by linear discriminant analysis (SPSS 15.0), Support Vector Machine (SVM), and Multilayer perceptron (MLP) using substracting baseline values from the emotional state. Results: The emotional stimuli had 96% validity and 5.8 effectiveness on average. The result of linear discriminant analysis using physiological signals showed that an accuracy of three different emotions classification was 83.4%. And an accuracy of three emotions classification by SVM was 75.5% and 55.6% by MLP. Conclusion: This study identified that three emotions were classified by linear discriminant analysis using various physiological features. Future study is needed to obtain stability and reliablity of this result compare with accuracy of emotion classification using other algorithms. Application: This could help emotion recognition studies lead to better chance to recognize various human emotions by using physiological signals as well as is able to be applied on human-computer interaction system for emotion reecognition.
- KSP Keywords
- Baseline values, Emotion Recognition, Emotional states, Human computer interaction, Human emotions, Interaction System, Physiological features, Physiological signals, Support VectorMachine(SVM), assessment scale, emotion classification