• 디자인융복합연구
• /
• 제16권6호
• /
• pp.137-152
• /
• 2017

본 연구는 실버세대를 위한 컴패니언 로봇의 인터랙션 경험 디자인을 위해 사용자 태스크- 로봇 기능 적합도 매핑에 기반한 로봇 유형 분석과 멀티모달 대화 인터랙션에서의 로봇 감정표현 연구를 수행하였다. 노인의 니즈 분석을 위해 노인과 자원 봉사자를 대상으로 FGI, 에스노그래피를 진행하였으며 로봇 지원 기능과 엑추에이터 매칭을 통해 로봇 기능 조합 유형에 대한 분석을 하였다. 도출된 4가지 유형의 로봇 중 표정 기반 대화형 로봇 유형으로 프로토타이핑을 하였으며 에크만의 얼굴 움직임 부호화 시스템(Facial Action Coding System: FACS)을 기반으로 6가지 기본 감정에 대한 표정을 시각화하였다. 사용자 실험에서는 로봇이 전달하는 정보의 정서코드에 맞게 로봇의 표정이 변화할 때와 로봇이 인터랙션 사이클을 자발적으로 시작할 때 사용자의 인지와 정서에 미치는 영향을 이야기 회상 검사(Story Recall Test: STR)와 표정 감정 분석 소프트웨어 Emotion API로 검증하였다. 실험 결과, 정보의 정서코드에 맞는 로봇의 표정 변화 그룹이 회상 검사에서 상대적으로 높은 기억 회상률을 보였다. 한편 피험자의 표정 분석에서는 로봇의 감정 표현과 자발적인 인터랙션 시작이 피험자들에게 정서적으로 긍정적 영향을 주고 선호되는 것을 확인하였다.

• 한국감성과학회:학술대회논문집
• /
• 한국감성과학회 1999년도 춘계학술발표논문집 논문집
• /
• pp.11-15
• /
• 1999

The discrete state theory on emotion postulated that there existed discrete emotions, such as happiness, anger, fear, disgust, and so forth. Many investigators who emphasized discreteness of emotions have suggested that discrete emotions entailed their specific activities in the autonomic nervous system. The purposes of this study were to develop a model of emotion-specific physiological response patterns. The study postulated six emotions (i.e., happiness, sadness, anger, disgust, fear, and surprise) as the basic discrete emotions. Thirty eight college students participated in the present study. Twelve slides (2 for each emotion category) were presented to the subjects in random order. During resting period of 30 s prior to the presentation of each slide, four presentation of each slide, four physiological measures (EEG, ECG, EDA, and respiration) were recorded to establish a baseline. The same physiological measures were recorded while each slide was being presented for 60 s (producing an emotional sate). Then, the subjects were asked to rate the degree of emotion induced by the slide on semantic differential scales. This procedure was repeated for every slide. Based upon the results, a model of emotion-specific physiological response patterns was developed: four emotion (fear, disgust, sadness, and anger) were classified according to the characteristics of EEG and autonomic responses. However, emotions of happiness and surprise were not distinguished by any combination of the physiological measures employed in this study, suggesting another appropriate measure should be adopted for differentiation.

• 한국음향학회지
• /
• 제25권8호
• /
• pp.375-382
• /
• 2006

본 논문에서는 음성신호에 포함된 감정정보를 자동으로 식별하는 방법과 특정 감정을 검출하는 방법에 대해 다룬다. 자동 감정식별 및 검출을 위해 장구간 (long-term) 음향 특징을 사용하였고, F-score 기반의 특징선택 기법을 적용하여 최적의 특징 파라미터들을 선정하였다. 기존의 일반적인 SVM을 확률출력 SVM으로 변환하여 감정식별 및 감정검출 시스템을 구축하였으며, 가설검정에 기반한 감정검출을 위해 세 가지의 대수 우도비 (log-likelihood) 근사법을 제안하여 그 성능을 비교하였다. SUSAS 데이터베이스를 사용한 실험 결과, F-score를 이용한 특징선택 기법에 의해 감정식별 성능이 향상되었으며, 확률출력 SVM의 유효성을 검증할 수 있었다. 감정검출의 경우, 제안한 방법에 의해 91.3%의 정확도로 화난 감정을 검출할 수 있었다.

• 디지털융복합연구
• /
• 제16권7호
• /
• pp.97-105
• /
• 2018

4차 산업혁명은 정보통신 분야에 큰 변화를 가져오고 그로 인해 다양한 일자리가 창출될 것으로 예상된다. 그러나 신기술 등장에 따른 직무 스트레스와 불확실성으로 IT 인력의 심리적 불안감은 증대되고 있다. 연구자는 전문직 정체성이 정보시스템(IS) 운영인력의 직무 스트레스와 정서적 소진을 어떻게 조절하는지 살펴보았다. 실증연구를 위해 정보시스템 관리 및 지원업무를 담당하는 160명을 대상으로 설문을 배포하여 데이터를 수집하였고, SPSS 21을 사용하여 가설을 검증하였다. 분석 결과, 직무 스트레스의 구성요소인 역할갈등, 역할모호, 질적 업무과부하는 정보시스템 운영인력의 정서적 소진에 영향을 미쳤으며, 전문직 정체성은 질적 업무과부하와 정서적 소진의 관계를 조절하고 있었다. 반면, 전문직 정체성은 역할갈등과 정서적 소진, 역할모호와 정서적 소진의 관계를 조절하지 않는 것으로 나타났다. 전문직 정체성이 신기술 도입에 따른 심리적 부담과 정서적 소진을 조절하는 만큼, 조직에서는 개개인의 전문직 정체성을 높이기 위한 지원을 제공해야 할 것이다.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2003년도 춘계학술대회
• /
• pp.935-939
• /
• 2003

The emotion recognition is one of the most significant interface technologies which make the high level of human-machine communication possible. The central nervous system stimulated by emotional stimuli affects the autonomous nervous system like a heart, blood vessel, endocrine organs, and so on. Therefore bio-signals like HRV, ECG and EEG can reflect one' emotional state. This study investigates the correlation between emotional states and bio-signals to realize the emotion recognition. This study also covers classification of human emotional states, selection of the effective bio-signal and signal processing. The experimental results presented in this paper show possibility of the emotion recognition.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2002년도 추계학술대회논문초록집
• /
• pp.375.2-375
• /
• 2002

This paper describes the emotion recognition algorithm using HMM(Hidden Markov Model) method. The relation between the mechanic system and the human has just been unilateral so far This is the why people don't want to get familiar with multi-service robots. If the function of the emotion recognition is granted to the robot system, the concept of the mechanic part will be changed a lot. (omitted)

• International journal of advanced smart convergence
• /
• 제9권1호
• /
• pp.193-201
• /
• 2020

The main objective of this study is to investigate the impact of additional modalities on the performance of emotion recognition using speech, facial expression and physiological measurements. In order to compare different approaches, we designed a feature-based recognition system as a benchmark which carries out linear supervised classification followed by the leave-one-out cross-validation. For the classification of four emotions, it turned out that bimodal fusion in our experiment improves recognition accuracy of unimodal approach, while the performance of trimodal fusion varies strongly depending on the individual. Furthermore, we experienced extremely high disparity between single class recognition rates, while we could not observe a best performing single modality in our experiment. Based on these observations, we developed a novel fusion method, called parametric decision fusion (PDF), which lies in building emotion-specific classifiers and exploits advantage of a parametrized decision process. By using the PDF scheme we achieved 16% improvement in accuracy of subject-dependent recognition and 10% for subject-independent recognition compared to the best unimodal results.

• 말소리와 음성과학
• /
• 제2권1호
• /
• pp.77-85
• /
• 2010

This paper proposes an efficient feature vector processing technique to guard the Speech Emotion Recognition (SER) system against a variety of noises. In the proposed approach, emotional feature vectors are extracted from speech processed by comb filtering. Then, these extracts are used in a robust model construction based on feature vector classification. We modify conventional comb filtering by using speech presence probability to minimize drawbacks due to incorrect pitch estimation under background noise conditions. The modified comb filtering can correctly enhance the harmonics, which is an important factor used in SER. Feature vector classification technique categorizes feature vectors into either discriminative vectors or non-discriminative vectors based on a log-likelihood criterion. This method can successfully select the discriminative vectors while preserving correct emotional characteristics. Thus, robust emotion models can be constructed by only using such discriminative vectors. On SER experiment using an emotional speech corpus contaminated by various noises, our approach exhibited superior performance to the baseline system.

• 감성과학
• /
• 제2권1호
• /
• pp.1-25
• /
• 1999

The differentiation of emotions by means of psychophysiological measures has been only moderately successful so far. A major reason for this dilemma may be the lack of appropriate neurophysiological modeling for the various autonomic nervous system based measures being used in emotion research. The aim of the present article is to provide such a neurophysiological background for electrodermal activity which has been frequently used as an indicator of emotional processes. First, the literature is reviewd with respect to the usability of electrodermal measures as an indicators of emotion. second, the neurophysilogical sources of electrodermal phenomena in general are described. Electrodermal activity has different origins in the central nervous system, a limbic-hypothalamic source that dominates during negative emotions as opposed to a premotor and basal ganglia source being predominantly active during positive emotions. Panksepp's model of four basic emotive systems is adopted for demonstrating subcortical structures and pathways possibly involved in the elicitation of both kinds of electrodermal activity in comparison with cardiovascular in dicators of emotional processes.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 추계학술발표대회
• /
• pp.528-531
• /
• 2021

Nowadays recommendation systems are so ubiquitous, where our many decisions are being done by the means of them. We can see recommendation systems in all areas of our daily life. Therefore the research of this sphere is still so active. So far many research papers were published for clothing recommendations as well. In this paper, we propose the clothing-recommendation system according to user emotion and weather information. We used social media to analyze users' 6 basic emotions according to Paul Eckman theory and match the colour of clothing. Moreover, getting weather information using visualcrossing.com API to predict the kind of clothing. For sentiment analysis, we used Emotion Lexicon that was created by using Mechanical Turk. And matching the emotion and colour was done by applying Hayashi's Quantification Method III.