• Journal of Korea Society of Industrial Information Systems
• /
• v.26 no.1
• /
• pp.1-10
• /
• 2021

This study proposed a Long-Short Term Memory network to consider changes in emotion over time, and applied an attention mechanism to give weights to the emotion states that appear at specific moments. We used 32 channel EEG data from DEAP database. A 2-level classification (Low and High) experiment and a 3-level classification experiment (Low, Middle, and High) were performed on Valence and Arousal emotion model. As a result, accuracy of the 2-level classification experiment was 90.1% for Valence and 88.1% for Arousal. The accuracy of 3-level classification was 83.5% for Valence and 82.5% for Arousal.

• Journal of Digital Convergence
• /
• v.15 no.3
• /
• pp.107-117
• /
• 2017

This study examined the conditional indirect effect of EEG (electroencephalogram) arousal on the relationship among affective valence, visual attention, perceived self-relevance, and attitudes toward campaign issues in the context of public service announcements (PSAs). Using SPSS macro (No. 14) of conditional process model, the findings in this current study indicated that the perceived self-relevance mediates the relationship between affective valence of PSA and attitudes toward issues and, in turn, is moderated by EEG arousal, indicating goodness-of-fit of the moderated mediation of psychophysiological arousal on PSAs. The results suggested that management of PSAs should be considered the strategic combination between affective valence and perceived self-relevance in advertising appeals.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.55 no.5
• /
• pp.240-247
• /
• 2006

Frequent arousals during sleep degrade the quality of sleep and result in sleep fragmentation. Visual inspection of physiological signals to detect the arousal events is cumbersome and time-consuming work. The purpose of this study is to develop an automatic algorithm to detect the arousal events. The proposed method is based on time-frequency analysis and the support vector machine classifier using single channel electroencephalogram (EEG). To extract features, first we computed 6 indices to find out the informations of a subject's sleep states. Next powers of each of 4 frequency bands were computed using spectrogram of arousal region. And finally we computed variations of power of EEG frequency to detect arousals. The performance has been assessed using polysomnographic (PSG) recordings of twenty patients with sleep apnea, snoring and excessive daytime sleepiness (EDS). We could obtain sensitivity of 79.65%, specificity of 89.52% for the data sets. We have shown that proposed method was effective for detecting the arousal events.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.37 no.1
• /
• pp.1-8
• /
• 2014

This paper proposes an emotion classifier from EEG signals based on Bayes' theorem and a machine learning using a perceptron convergence algorithm. The emotions are represented on the valence and arousal dimensions. The fast Fourier transform spectrum analysis is used to extract features from the EEG signals. To verify the proposed method, we use an open database for emotion analysis using physiological signal (DEAP) and compare it with C-SVC which is one of the support vector machines. An emotion is defined as two-level class and three-level class in both valence and arousal dimensions. For the two-level class case, the accuracy of the valence and arousal estimation is 67% and 66%, respectively. For the three-level class case, the accuracy is 53% and 51%, respectively. Compared with the best case of the C-SVC, the proposed classifier gave 4% and 8% more accurate estimations of valence and arousal for the two-level class. In estimation of three-level class, the proposed method showed a similar performance to the best case of the C-SVC.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.6
• /
• pp.891-898
• /
• 2012

The study for the passenger's comfortableness of vehicles and the arousal of car drivers has been done widely. On the other hand, there are few studies for the locomotive engineers. Human error means that the mistakes made by human, recently it receives attention in the field of safety engineering and human engineering. Comparing the operating condition of train with car, because of the simplification of the visual stimulus, the arousal level on the train goes down easily. The arousal level down makes judgement down, the accident risk from human error is getting bigger. In this study, we measured bio-signals(ECG, EDA, PPG, respiration and EEG) from 6 locomotive engineers to evaluate their arousal state while they operated the train. Also we recorded the 3 axes acceleration signal showing the vibration state of train. Also, the existence of tunnels were simultaneously measured. At the station section where the train speed goes down, the size of vector's sum decreases because of reduced vibration. Beta component in EEG tends to increase at the entering point of each station and tunnel. It is due to the arousal reaction and tension growth. The mean SCR(skin conductance response) was more increased in neutral section. As the button control movement (body movement) increases in the neutral section, it is appeared that SCR increase. RR interval tends to gradually increase during train operation for 1 hour 40 minutes. However, It tends to sharply decrease at the stop station because strong concentration needed to stop train on the exact point. The engineer's arousal reaction can be checked through analysing the bio-signal change during train operation. Therefore, if this analysing result is adopted to the sleepiness prevention caution system, it will be useful for the safety train operation.

• Journal of the Ergonomics Society of Korea
• /
• v.21 no.3
• /
• pp.35-46
• /
• 2002

In this study, emotion changes were induced by four imaginations- pleasantness, unpleasantness, arousal, relaxation and it was examined using subjective evaluation and analysis of the physiological signals of the central and autonomic nerve systems whether the intended emotions were appropriately achieved, and whether these emotion changes could be distinguished from the analysis of physiological signals. Each of the four imaginations was implemented on 32 subjects for 30 seconds, while that Electroencephalogram (EEG), Eelectrocardiogram (RSP) were measured, and a subjective evaluation was implemented following the completion of the measurement. The analysis of the subjective evaluation revealed that the subjects underwent the four clearly differentiated imaginations, and the pleasantness level was classified into four imagination stages, pleasantness>relaxation>arousal=comfort>unpleasantness, and arousal level was classified into four imagination stages in the order of arousal>unpleasantness${\approx}$pleasantness>comfort>relaxation. The analysis of the EEG revealed that three stages of pleasantness level, pleasantness>relaxation=arousal=comfort>unpleasantness were classified from the values of ${\alpha}/{\alpha}+{\beta}\;and\;{\beta}/{\alpha}+{\beta}$, and about tour distinguishable stages of arousal level were obtained from the autonomic nervous system responses following the order of arousal>unpleasantness${\approx}$pleasantness> comfort> relaxation. It was found that intended emotion could be induced from the imagination, and these induced emotion changes could be differentiated using the physiological signals of the EEG and autonomic nervous system.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.35 no.2
• /
• pp.64-70
• /
• 2012

This paper presents a statistical analysis method for the selection of electroencephalogram (EEG) electrode positions and spectral features to recognize emotion, where emotional valence and arousal are classified into three and two levels, respectively. Ten experiments for a subject were performed under three categorized IAPS (International Affective Picture System) pictures, i.e., high valence and high arousal, medium valence and low arousal, and low valence and high arousal. The electroencephalogram was recorded from 12 sites according to the international 10~20 system referenced to Cz. The statistical analysis approach using ANOVA with Tukey's HSD is employed to identify statistically significant EEG electrode positions and spectral features in the emotion recognition.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.819-820
• /
• 2006

The purpose of this study is to develop an automatic algorithm to detect the arousal events. The proposed method is based on time-frequency analysis and the support vector machine classifier using single channel electroencephalogram. To extract features, first we computed 6 indices to find out the information of sleep states. Next powers of each of 4 frequency bands were computed using spectrogram of arousal region. And finally we computed variations of power of EEG frequency to detect arousals. The performance has been assessed using polysomnographic recordings of twenty patients with sleep apnea, snoring and excessive daytime sleepiness. We have shown that proposed method was effective for detecting the arousal events.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.10
• /
• pp.451-456
• /
• 2014

The correlations between electroencephalographic (EEG) spectral power and emotional responses during affective sound clip listening are important parameters. Hemispheric asymmetry in prefrontal activation have been proposed in two decades ago, as measured by power value, is related to reactivity to affectively pleasure audio stimuli. In this study, we designed an emotional audio stimulus experiment in order to verify frontal EEG asymmetry by analyzing Event-related Spectral Perturbation (ERSP) results. Thirty healthy college male students volunteered the stimulus experiment with the standard IADS(International Affective Digital Sounds) clips. These affective sound clips are classified in three emotion states, high pleasure-high arousal (happy), middle pleasure-low arousal (neutral) and low pleasure-high arousal (fear). The analysis of the data was performed in both theta (4-8Hz) and alpha (8-13Hz) bands. ERSP maps in the alpha band revealed that there are the stronger power responses of high pleasure (happy) in the right frontal lobe, while the stronger power responses of middle-low pleasure (neutral and fear) in the left frontal lobe. Moreover, ERSP maps in the theta band revealed that there are the stronger power responses of high arousal (fear and happy) in the left pre-frontal lobe, while the stronger responses of low arousal (neutral) in the right pre-frontal lobe. However, the high pleasure emotions (happy) can elicit greater relative right EEG activity, while the low and middle pleasure emotions (fear and neutral) can elicit the greater relative left EEG activity. Additionally, the most differences of theta band have been found out in the medial frontal lobe, which is proved as the frontal midline theta. And there are the strongest responses of happy sounds in the alpha band around the whole frontal regions. These results are well suited for emotion recognition, and provide the evidences that theta and alpha powers may have the more important role in the emotion processing than previously believed.

• Journal of Sensor Science and Technology
• /
• v.4 no.3
• /
• pp.30-36
• /
• 1995

One of principal causes of car accidents is low arousal level of driver. Drivers arrive their destination under an appropriate arousal level. Basic research was done to develop a portable arousal monitor and to evaluate the arousal level of drivers. An arousal monitor which can simulantaneously measure the skin impedance change and EEG was designed and tested. The relationship among three parameters was studied and was used to determine the index of skin impedance level depending on arousal level.