• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.4
• /
• pp.153-160
• /
• 2021

With the advent of the "age of video" due to the simplification of video content production and the convenience of broadcasting channel operation, review videos on various products are drawing attention. We proposes RASIA, an integrated reliability analysis system based on verbal and nonverbal sentiment analysis of review videos. RASIA extracts and quantifies each emotional value obtained through language sentiment analysis and facial analysis of the reviewer in the video. Subsequently, we conduct an integrated reliability analysis of standardized verbal and nonverbal sentimental values. RASIA provide an new objective indicator to evaluate the reliability of the review video.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.313-316
• /
• 2020

감성 분석은 문장의 감성을 분석해 긍정 또는 부정으로 분류하는 작업을 의미한다. 문장에 담긴 감성을 파악해야 하기 때문에 문장 전체를 이해하는 것이 중요하다. 그러나 한 문장에 긍정과 부정의 이중 극성이 동존하는 문장은 감성 분석에 혼동이 생길 수 있다. 본 논문에서는 이와 같은 문제를 해결하기 위해 단어의 감성 점수 예측을 통해 감성 단어 등장 순서를 고려한 감성 분석 모델을 제안한다. 또한 최근 다양한 자연어 처리 분야에서 좋은 성능을 보이는 사전 학습 언어 모델을 활용한다. 실험 결과 감성 분석 정확도 90.81%로 기존 모델들에 비해 가장 좋은 성능을 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.515-518
• /
• 2020

영상 콘텐츠 생산 간편화와 방송 채널 운영의 편리화에 따른 '영상의 시대'가 도래함에 따라 여러 제품에 대한 리뷰 영상이 관심을 받고 있다. 본 연구에서는 리뷰 영상의 언어와 비언어적 감성 분석을 토대로 통합 신뢰도 분석 시스템을 제안한다. 이를 위해, 영상 속 음성의 언어 감성 분석과 리뷰어의 표정 분석을 통해 얻은 각 감성값을 추출하고 정량화한다. 이후 표준화된 언어, 비언어적 감성 값에 대한 통합 신뢰도 분석을 진행한다. 결과적으로, 리뷰 영상에 대한 신뢰도를 객관화된 지표로써 평가할 수 있다.

• Science of Emotion and Sensibility
• /
• v.4 no.2
• /
• pp.15-24
• /
• 2001

This study is designed to examine on the effectiveness of emotional advertisement, including analysis of the physiological response of customers’ feelings for effective development of advertisement, and to provide a base for emotional advertisement production by presenting the methods of linguistic and non-linguistic response. It provides a framework for symbolic analysis about costumers’ emotions along with a general review of emotional advertisement. Based on the results, it will analyze aspects of linguistic and non-linguistic utterance of costumers’ physiological response for emotional advertisement. The results will present a frame of effective emotional advertisement that appeals to the emotion of costumers

• Science of Emotion and Sensibility
• /
• v.17 no.4
• /
• pp.61-70
• /
• 2014

Verbal communication has limits in the interaction between robot and man, and therefore nonverbal communication is required for realizing smoother and more efficient communication and even the emotional expression of the robot. This study derived 7 pieces of nonverbal information based on shopping behavior using the robot designed to support shopping, selected facial expression as the element of the nonverbal information derived, and coded face components through 2D analysis. Also, this study analyzed the significance of the expression of nonverbal information using 3D animation that combines the codes of face components. The analysis showed that the proposed expression method for nonverbal information manifested high level of significance, suggesting the potential of this study as the base line data for the research on nonverbal information. However, the case of 'embarrassment' showed limits in applying the coded face components to shape and requires more systematic studies.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.57-67
• /
• 2024

This research investigates training methods for large language models to accurately identify sentiments and comprehend information about increasing and decreasing fluctuations in the financial domain. The main goal is to identify suitable datasets that enable these models to effectively understand expressions related to financial increases and decreases. For this purpose, we selected sentences from Wall Street Journal that included relevant financial terms and sentences generated by GPT-3.5-turbo-1106 for post-training. We assessed the impact of these datasets on language model performance using Financial PhraseBank, a benchmark dataset for financial sentiment analysis. Our findings demonstrate that post-training FinBERT, a model specialized in finance, outperformed the similarly post-trained BERT, a general domain model. Moreover, post-training with actual financial news proved to be more effective than using generated sentences, though in scenarios requiring higher generalization, models trained on generated sentences performed better. This suggests that aligning the model's domain with the domain of the area intended for improvement and choosing the right dataset are crucial for enhancing a language model's understanding and sentiment prediction accuracy. These results offer a methodology for optimizing language model performance in financial sentiment analysis tasks and suggest future research directions for more nuanced language understanding and sentiment analysis in finance. This research provides valuable insights not only for the financial sector but also for language model training across various domains.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.12
• /
• pp.49-56
• /
• 2014

In this paper, we propose a method for identifying hidden principal sentiments among large scale texts from documents, social data, internet and blogs by analyzing standard language, slangs, argots, abbreviations and emoticons in those words. The IRLBA(Implicitly Restarted Lanczos Bidiagonalization Algorithm) is used for principal component analysis with large scale sparse matrix. The proposed system consists of data acquisition, message analysis, sentiment evaluation, sentiment analysis and integration and result visualization modules. The suggested approaches would help to improve the accuracy and expand the application scope of sentiment analysis in social data.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.586-590
• /
• 2023

인터넷과 소셜미디어 사용량의 급증으로, 제품 리뷰, 온라인 피드백, 소셜 미디어 게시물 등을 통해 고객의 감정을 파악하는 것이 중요해졌다. 인공지능이 활용되어 고객이 제품이나 서비스의 어떤 부분에 만족하거나 불만을 가지는지를 분석하는 연구를 ABSA라고 하며 이미 해외에서는 이런 연구가 활발하게 이루어지는 반면, 국내에서는 상대적으로 부족한 상황이다. 이 연구에서는 ABSA의 두 개의 주요 작업인 ACD와 ASC에 대해 생성 모델 중 하나인 T5 모델을 사용하는 방법론을 제시한다. 이 방법론은 기존 판별 모델을 사용하는 것에 비해 시간과 성능 측면에서 크게 향상되었음을 보여준다.

• Journal of Internet of Things and Convergence
• /
• v.4 no.2
• /
• pp.29-46
• /
• 2018

The purpose of this study is to understand the relation among experience of verbal abuse, emotional labor, emotional intelligence, social support and turnover intention of hospital nurses. The survey was conducted with 189 hospital nurses working at general hospital in the city of S. The data was collected by using structured questionnaires as research tools from November 17, 2015 to November 27, 2015. The collected data was analyzed by using SPSS ver. 18.0 program. As a result of the correlation analysis between the nurse 's experience of verbal abuse, emotional labor, emotional intelligence, social support and turnover intention, the turnover intention of the nurse was a significantly positive correlation between the experiences of verbal abuse from the doctor, nurses, patients, and the caregiver and the emotional labor. Also, There is a significantly negative correlation between emotional intelligence and social support. In order to identify the factors influencing the turnover intention of the subjects, multiple regression analysis was performed. The statistically significant variables were emotional labor & the ER in the working department.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.3
• /
• pp.9-15
• /
• 2019

Large global online shopping malls, such as Amazon, offer services in English or in the language of a country when their products are sold. Since many customers purchase products based on the product reviews, the shopping malls actively utilize the sentimental analysis technique in judging preference of each product using the large amount of review data that the customer has written. And the result of such analysis can be used for the marketing to look the potential shoppers. However, it is difficult to apply this English-based semantic analysis system to different languages used around the world. In this study, more than 500,000 data from Amazon fine food reviews was used for training a deep learning based system. First, sentiment analysis evaluation experiments were carried out with three models of English test data. Secondly, the same data was translated into seven languages (Korean, Japanese, Chinese, Vietnamese, French, German and English) and then the similar experiments were done. The result suggests that although the accuracy of the sentimental analysis was 2.77% lower than the average of the seven countries (91.59%) compared to the English (94.35%), it is believed that the results of the experiment can be used for practical applications.