• 전기전자학회논문지
• /
• 제24권4호
• /
• pp.998-1003
• /
• 2020

본 논문에서는 기존 감정 인식 시스템의 성능 향상을 위하여 감정 적응을 사용한 감정 학습 방법이 제안되었다. 감정 적응을 위하여 적은 개수의 학습 감정 음성과 감정 적응 방식을 사용하여 감정이 없는 음성 모델로부터 감정 음성 모델이 생성되었다. 이러한 방법은 기존 방법보다 적은 개수의 감정 음성을 사용하여도 우수한 성능을 나타내었다. 학습을 위하여 충분한 감정 음성을 얻는 것은 쉽지 않기 때문에 적은 개수의 감정 음성을 사용하는 것은 실제 상황에서 매우 실용적이다. 4가지 감정이 포함된 한국어 데이터베이스를 사용한 실험 결과에서 감정 적응을 이용한 제안된 방법이 기존 방법보다 우수한 성능을 나타내었다.

• 대한통합의학회지
• /
• 제6권2호
• /
• pp.69-76
• /
• 2018

Purpose : This study investigated the effects of low-intensity muscle training on neurotransmitters associated with depression and emotion regulation disorders in survivors with stroke who were diagnosed with depression or emotion regulation disorder. Methods : Survivors of chronic stroke diagnosed with depression or emotion regulation disorder (n = ) were included in the study. Concentrations of serotonin, norepinephrine, and epinephrine were measured before and after. Low-intensity muscle training was performed once a day for 50 minutes, 5 times a week for 10 weeks. Result : Levels of serotonin, norepinephrine, and epinephrine improved significantly after intervention (p < .05). Conclusion : These results confirm that low-intensity muscle training has a positive effect on neurotransmitter concentrations in survivors with stroke diagnosed with depression and emotion regulation disorders. Future studies should analyze the changes in neurotransmitters in more detail using various exercise methods and intensities to investigate their effects on depression and emotion regulation disorders.

• 한국학교보건학회지
• /
• 제30권3호
• /
• pp.295-305
• /
• 2017

Purpose: This study was conducted to evaluate the effect of Mindfulness-based Emotion Management Training for adolescents immersed in self-harm behaviors. Methods: The participants were 61 middle school students located in Seoul. Thirty students were assigned to the Mindfulness-based Emotion Management Training (M-EMT) group and the other 31 students were assigned to the waitlist group. The students in the M-EMT group participated in a total of 12 sessions, twice a week, for 6 weeks. The modules of the program comprised of mindfulness for emotion, mindfulness for behavior/impulse, mindfulness for thought, and coping with self-harming behavior. The Distress Tolerance Scale, Deliberate Self-Harm Inventory, Cognitive Emotion Regulation Questionnaire, and Scale Suicidal Ideation were assessed at baseline and post-intervention. Results: Compared to the waitlist group, the intervention group exhibited a statistically significant improvement in distress tolerance and cognitive emotion regulation and a statistically significant reduction in self-harming behaviors and suicidal ideation. Conclusion: These results support the hypothesis that mindfulness-based emotional management training has a positive effect on reducing self-harming behaviors and enhancing distress tolerance and cognitive emotional regulation.

• 한국컴퓨터그래픽스학회논문지
• /
• 제24권4호
• /
• pp.1-9
• /
• 2018

본 논문은 얼굴의 표정 변화를 통해 감정을 분석하는 방법으로 조현병의 초기 증상을 스스로 인지할 수 있는 감정 트레이닝 프레임워크를 제안한다. 먼저, Microsoft의 Emotion API를 이용하여 캡처된 얼굴 표정의 사진으로부터 감정값을 얻고, 피크 분석 기반 표준편차로 시간에 따라 변화하는 얼굴 표정의 미묘한 차이를 인식해 감정 상태를 각각 분류한다. 그리하여 Ekman이 제안한 여섯 가지 기본 감정 상태에 반하는 감정들의 정서 및 표현능력이 결핍된 부분에 대해 분석하고, 그 값을 이미지 색상 변환 프레임워크에 통합시켜 사용자 스스로 감정의 변화를 쉽게 인지하고 트레이닝 할 수 있도록 하는 것이 최종목적이다.

• 한국감성과학회:학술대회논문집
• /
• 한국감성과학회 2012년도 춘계학술대회
• /
• pp.132-133
• /
• 2012

• 대한불안의학회지
• /
• 제18권1호
• /
• pp.10-16
• /
• 2022

Objective : The present study examined the effectiveness of the mindful self-compassion (MSC) program on anxiety, depression, self-compassion and emotion regulation. Methods : A total of 29 subjects (mean age 27.5±6.5 years, 15 males and 14 females) participated in a standardized 8-week MSC program. The control group consisted of age- and sex-matched twenty participants (mean age 26.0±2.9 years, 11 males and 9 females). All subjects completed self-report measurements at two weeks before and after the MSC program. Results : MSC training improved self-compassion as demonstrated by the significant group x time interaction effects on the total Self-Compassion Scale scores (F[1, 47]=8.324, p<0.01). Regarding the subscale scores, a significant improvement in self-kindness, isolation and mindfulness components of self-compassion was observed after MSC training. A significant group x time interaction was observed on the self-kindness subscale (F[1, 47]=4.664, p<0.05), with a significant main effect of time (F[1, 47]=23.723, p<0.001). The isolation subscale showed a significant group x time interaction (F[1, 47]=8.698, p<0.001). For the mindfulness subscale, there was a significant group x time interaction (F[1, 47]=6.611, p<0.05) and main effect of time (F[1, 47]=6.611, p<0.05). MSC training also improved the acceptance emotion regulation strategy, as demonstrated by the significant group x time interaction in the acceptance subscale scores of the Cognitive Emotion Regulation Questionnaire (F[1, 47]=6.845, p<0.05). Conclusion : MSC training showed efficacy in fostering self-compassion and improving emotion regulation. Thus, this program might be applicable to improve mental health.

• 아동학회지
• /
• 제33권6호
• /
• pp.1-19
• /
• 2012

The purpose of this study was to develop an intervention program for the improvement of the emotion regulation of 5-year-olds and to analyze the effects of this program on their emotion regulation and social behavior. The participants consisted of 56 5-year-olds from two day care center located in a metropolitan area of Korea. The experimental group (n=28) participated in the intervention program, and the comparative group (n=28) did not. This intervention program was developed to deal with peer relationships, and consists of activities designed to cover all domains of the Korean National Early Education Standard. The emotion regulation and social behavior were measured before and after the experiment. The data were analyzed making use of ANCOVA. Our results indicate that this program had the effect of improving the emotion regulation and social behavior on the part of 5-year-olds.

• 대한물리의학회지
• /
• 제7권1호
• /
• pp.125-135
• /
• 2012

Purpose : This study is designed to demonstrate the effects of circuit obstacle group gait training on walking ability and emotion in stroke patients. Methods : Twenty one patients with stroke were participated in this study. The subjects were divided into control group(n=10) and experimental group(n=11). Circuit obstacle group gait training consisted of walking around obstacles, walking over obstacles, walking up and down slopes and walking up and down stairs. Circuit obstacle group gait training was conducted five times per week, 1 hour per session, for 6 consecutive weeks. At pre-test and post-test, subjects were tested with 10 m walking test, timed up and go test, up and down 4 stairs test, depression and self esteem. Results : After 6 weeks of research, the experimental group showed statistically significant difference in all items when comparing prior to training and after training (p<.05), but the control group showed statistically significant difference in items other than depression and self esteem(p<.05). In the comparison between the two groups, the experimental group showed higher improvement than the control group in the 10 m walking test, timed up and go test, and up and down 4 stairs test, and there was statistically significant difference in decrease of degree in depression between the experimental group and control group(p<.05). Conclusion : This study have shown that circuit obstacle group gait training improves walking ability and emotion in stroke patients.

• 전기전자학회논문지
• /
• 제22권3호
• /
• pp.816-821
• /
• 2018

최근 감정 분석 및 질병 진단을 위한 뇌파 연구 분야에서 인공 신경망을 기반으로 한 기계학습 알고리즘이 분류기로 널리 사용되기 시작했다. 뇌파 데이터 분류를 위해 기계학습 모델을 사용하는 경우 유사한 특성을 가지는 데이터만으로 학습데이터가 구성되면 다른 그룹의 데이터에 적용했을 때 분류 성능이 떨어질 수 있다. 본 논문에서는 이러한 문제점을 개선하기 위해 준 지도학습 알고리즘을 사용해 여러 그룹의 데이터를 선택하여 학습데이터 세트를 구성하는 방법을 제안한다. 이후 제안하는 방법을 사용하여 구성한 학습데이터 세트와 유사한 특성을 가지는 데이터로 구성된 학습데이터 세트로 모델을 학습하여 두 모델의 성능을 비교하였다.

• 한국음향학회지
• /
• 제40권5호
• /
• pp.515-522
• /
• 2021

음성감정인식을 위한 훈련 데이터는 감정 레이블링의 어려움으로 인해 충분히 확보하기 어렵다. 본 논문에서는 음성감정인식의 성능 개선을 위해 트랜스포머 기반 모델에 대규모 음성인식용 훈련 데이터를 통한 전이학습을 적용한다. 또한 음성인식과의 다중작업학습을 통해 별도의 디코딩 없이 문맥 정보를 활용하는 방법을 제안한다. IEMOCAP 데이터 셋을 이용한 음성감정인식 실험을 통해, 가중정확도 70.6 % 및 비가중정확도 71.6 %를 달성하여, 제안된 방법이 음성감정인식 성능 향상에 효과가 있음을 보여준다.