• 한국정보기술학회논문지
• /
• 제17권9호
• /
• pp.123-131
• /
• 2019

최근 소셜 네트워크 서비스 사용자들은 게시글을 통해 사회적 이슈 및 관심 콘텐츠들에 대한 자신의 감정을 적극적으로 표현하고 공유한다. 그 결과 소셜 네트워크에서의 개인 및 특정 집단의 감정 공유는 빠르게 확산된다. 그러므로 사용자들의 게시글에 대한 감정 분석 연구가 활발히 진행되고 있다. 그렇지만 다양한 감정이 포함된 게시글에 대한 감정 분석 연구가 미흡하다. 따라서 본 논문에서는 해시태그와 이미지를 이용한 인스타그램 게시글의 대표 감정 분석 방법을 제안한다. 이를 통해 사용자 게시글에 포함된 다종의 리소스를 활용하여 다중의 감정으로부터 대표 감정을 추출할 수 있으며 66.4%의 정확도와 81.7%의 재현율로 기존 방법보다 감정 분류 성능 향상을 보인다.

• 한국IT서비스학회지
• /
• 제16권4호
• /
• pp.149-160
• /
• 2017

Sentiment analysis is the process of determining whether a piece of document, text or conversation is positive, negative, neural or other emotion. Sentiment analysis has been applied for several real-world applications, such as chatbot. In the last five years, the practical use of the chatbot has been prevailing in many field of industry. In the chatbot applications, to recognize the user emotion, sentiment analysis must be performed in advance in order to understand the intent of speakers. The specific emotion is more than describing positive or negative sentences. In light of this context, we propose deep learning models for conducting multi-class sentiment analysis for identifying speaker's emotion which is categorized to be joy, fear, guilt, sad, shame, disgust, and anger. Thus, we develop convolutional neural network (CNN), long short term memory (LSTM), and multi-layer neural network models, as deep neural networks models, for detecting emotion in a sentence. In addition, word embedding process was also applied in our research. In our experiments, we have found that long short term memory (LSTM) model performs best compared to convolutional neural networks and multi-layer neural networks. Moreover, we also show the practical applicability of the deep learning models to the sentiment analysis for chatbot.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권2호
• /
• pp.810-831
• /
• 2019

Recently, continuous dimensional emotion recognition from audiovisual clues has attracted increasing attention in both theory and in practice. The large amount of data involved in the recognition processing decreases the efficiency of most bimodal information fusion algorithms. A novel algorithm, namely the incomplete Cholesky decomposition based kernel cross factor analysis (ICDKCFA), is presented and employed for continuous dimensional audiovisual emotion recognition, in this paper. After the ICDKCFA feature transformation, two basic fusion strategies, namely feature-level fusion and decision-level fusion, are explored to combine the transformed visual and audio features for emotion recognition. Finally, extensive experiments are conducted to evaluate the ICDKCFA approach on the AVEC 2016 Multimodal Affect Recognition Sub-Challenge dataset. The experimental results show that the ICDKCFA method has a higher speed than the original kernel cross factor analysis with the comparable performance. Moreover, the ICDKCFA method achieves a better performance than other common information fusion methods, such as the Canonical correlation analysis, kernel canonical correlation analysis and cross-modal factor analysis based fusion methods.

• 한국감성과학회:학술대회논문집
• /
• 한국감성과학회 2012년도 춘계학술대회
• /
• pp.1-2
• /
• 2012

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제5권1호
• /
• pp.1-6
• /
• 2005

In this paper, we propose an emotion recognition method using the facial images and speech signals. Six basic emotions including happiness, sadness, anger, surprise, fear and dislike are investigated. Facia] expression recognition is performed by using the multi-resolution analysis based on the discrete wavelet. Here, we obtain the feature vectors through the ICA(Independent Component Analysis). On the other hand, the emotion recognition from the speech signal method has a structure of performing the recognition algorithm independently for each wavelet subband and the final recognition is obtained from the multi-decision making scheme. After merging the facial and speech emotion recognition results, we obtained better performance than previous ones.

• 대한가정학회지
• /
• 제50권3호
• /
• pp.35-50
• /
• 2012

This study aims to investigate the effect of group home adolescents' emotion regulation and school adjustment. A survey was carried out on a total of 246 middle and high school students, who live in group homes. For data analysis, t-test, two-way ANOVA, Pearson's correlation analysis, multiple regression analysis, and reliability coefficients were carried out by using SPSS program(version 18.0). The findings of this study were as follows: First, there were significant differences in emotion regulation ability of group home adolescents, depending on the grade and gender. Second, there was an interaction by grade and gender in school adjustment of group home adolescents. Last, group home adolescents' gender, grade, and emotion regulation ability affected their school adjustment.

• 대한가정학회지
• /
• 제49권10호
• /
• pp.103-113
• /
• 2011

This study investigated the structural relationships between attachment, social competence, and emotion regulation. A survey was administered to 233 children of elementary school age(5th-6th grades) in G-city, Korea, using the IPPA-R, the Social Competence Inventory and an Emotion Regulation Scale. The collected data were then analyzed using a Canonical Correlation Analysis. First, the relationship between attachment and social competence was analyzed. The results showed that attachment and social competence have a positively correlated relationship. Peer attachments strongly affect the attributes of social competence when using a canonical variate analysis. Secondly, the relationship between attachment and emotional regulation was analyzed. The results showed that attachment and emotion regulation are also positively correlated. Maternal attachment particularly strongly affected the attributes of emotion regulation. Thirdly, the relationship between social competence and emotional regulation was analyzed. The results showed that social competence and emotional regulation have a positive relationship.

• 한국의류학회지
• /
• 제26권5호
• /
• pp.663-674
• /
• 2002

The objectives of this study were 1) to ascertain whether there was a difference in product performance (expressive or instrumental), which consumer recognized after using, related to consumer involvement toward clothing, 2) to clarify the effect of product performance on consumption emotion(positive or negative), 3) to investigate the effect of consumption emotion on satisfaction, and 4) to find out whether product performance had a direct effect on satisfaction toward product. The study was conducted in three steps. Through the two steps, measurement instruments were developed. At the last step, judgement sampling method were utilized to collect the data and subjects were 614 university students. Confirmatory factor analysis and structural equation model analysis were used to analyze the data. The results were as follows: 1) Consumer involvement had an effect on product performances but it was related to the expressive product performance more than to the instrumental product performance. 2) Product performance had positive influence on positive consumption emotion, while it had negative influence on negative consumption emotion. The results revealed that there were significant relationships between product performance and consumption emotion. 3) Positive consumption emotion had a positive effect on consumer satisfaction, on the other hand negative consumption emotion had a negative effect on consumer satisfaction. 4) Although the direct effects of product performances on satisfaction were larger than the indirect effects, product performance was greatly influential in consumption emotion and consumption emotion was strongly related to consumer satisfaction. Therefore, consumption emotion is an important determinant variable in the process of consumer satisfaction.

• 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.

• Asian Journal of Business Environment
• /
• 제5권4호
• /
• pp.13-18
• /
• 2015

Purpose - Based on previous studies, this study extends current research and investigates whether the sub-factors of emotional intelligence increase job satisfaction or employee intrinsic and extrinsic motivation and perceived organizational support. Research design, data, and methodology - This study categorizes service employees' (consultants) emotional intelligence into four sub-factors: regulation of emotion, appraisal of emotion, utilization of emotion, and expression of emotion. The study then investigates the sub-factor effects on job satisfaction. A total of 353 valid questionnaires were collected. Results - The results of the path analysis showed that appraisal, utilization, and expression of emotion had a positive effect on intrinsic motivation, and utilization of emotion had a positive effect on extrinsic motivation. Extrinsic motivation had a positive effect on perceived organizational support and job satisfaction, and perceived organizational support had a positive effect on job satisfaction. Conclusion - As consultants' utilization of emotion is rendered as the ability to use emotion to improve performance, the conclusion is that such factors as monetary performance incentives are important in order to boost job satisfaction of the consultants.