• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.507-511
• /
• 2020

오픈 도메인 대화에서 텍스트에 나타난 태도나 성향과 같은 화자의 주관적인 감정정보를 분석하는 것은 사용자들에게서 풍부한 응답을 이끌어 내고 동시에 제공하는 목적으로 사용될 수 있다. 하지만 한국어 감성분석에서 기존의 대부분의 연구들은 긍정과 부정 두개의 클래스 분류만을 다루고 있고 이는 현실 화자의 감정 정보를 정확하게 분석하기에는 어려움이 있다. 또한 최근에 오픈한 다중클래스로된 한국어 대화 감성분석 데이터셋은 중립 클래스가 전체 데이터셋의 절반을 차지하고 일부 클래스는 사용하기에 매우 적은, 다시 말해 클래스 간의 데이터 불균형 문제가 있어 다루기 굉장히 까다롭다. 이 논문에서 우리는 일곱개의 클래스가 존재하는 한국어 대화에서 세션들을 효율적으로 분류하는 기법들에 대해 논의한다. 우리는 극심한 클래스 불균형에도 불구하고 76.56 micro F1을 기록하였다.

• Annual Conference on Human and Language Technology
• /
• 2008.10a
• /
• pp.58-62
• /
• 2008

인터넷 상에서 상품을 구입할 때 고려하는 부분 중의 하나가 상품평이다. 하지만 이러한 상품평들을 개인이 일일이 확인 하는데에는 상당한 시간이 소요된다. 이러한 문제점을 줄이기 위해서 본 논문에서는 인터넷 상의 상품평에 대한 의견을 긍정, 부정, 일반으로 나누는 시스템을 제안한다. 제안 시스템은 CRFs 기계학습모델을 기반으로 하며, 연결어미, 형태소 유니그램, 슬라이딩 윈도우 기법의 형태소 바이그램을 자질로 사용한다. 실험을 위해서 가격비교 사이트의 모니터 카테고리에서 561개의 상품평을 수집하였다. 이 중 465개의 상품평을 학습 문서로 사용하였고 96개의 상품평을 실험 문서로 사용하였다. 제안 시스템은 실험결과 79% 정도의 정확도를 보였다. 추가 실험으로 제안 시스템이 사람들과 얼마나 비슷한 성능을 보이는지 알아보기 위해서 카파 테스트를 실시하였다. 카파 테스트를 실시한 결과, 사람간의 카파 계수는 0.6415였으며, 제안 시스템과 사람 간의 카파 계수는 평균 0.5976이였다. 결론적으로 제안 시스템이 사람보다는 떨어지지만 유사한 정도의 성능을 보임을 알 수 있었다.

• Journal of Korean Academy of Nursing
• /
• v.27 no.4
• /
• pp.871-879
• /
• 1997

인간은 사회적 접촉을 필요로 하고 타인과의 상호작용을 원하는 사회적 동물이다. 외로움은 한 개인의 사회적 관계의 망이 양적이나 질적으로 부족할 때 발생하는 불유쾌한 감정이다. 외로움은 매우 보편적인 감정이나 고통스러울 수 있으며 이러한 경험이 지속되면 정신적, 신체적 건강을 해칠 수 있을 뿐 아니라 자살을 초래하기도 한다. 특히 노인은 여러 가지 종류의 변화와 상실을 경험하기 때문에 외로움을 더 느낄 수 있다. 본 연구의 목적은 번역과 역번역 과정을 통해 Revised UCLA Loneliness Scale에 대한 한국어 UCLA 외로움 사정도구를 만들고 이 도구를 한국노인에게 적용하여 신뢰도와 타당도를 조사하는 것이다. 개정판UCLA외로움 사정 도구는 각각 3번씩의 번역과 역번역 과정을 통해 원본 영문도구와 역번역된 영문도구 사이에 의미의 차이가 없다고 판단될 때까지 만들어졌다. 번역과 역번역 과정은 미국에서 석사나 박사학위를 취득한 6명의 한국유학생들에 의해 이루어 졌으며 원본 영문도구와 역번역된 도구사이의 의미의 차이는 본 저자와 미국인 간호학 교수에 의해 사정되었다. 번역된 한국어 외로움 사정도구는 미국의 3개 주에 거주하는 203명의 한국노인들을 대상으로 신뢰도와 타당도 조사가 이루어졌다. 한국어 UCLA외로움 사정도구의 신뢰계수는 .93으로 한국어 도구의 내용이 전반적으로 동질성을 띈다는 것을 알 수 있었다. 이는 Barren등(1992)이 미국인 노인여성을 대상으로 한 연구에서 얻은 신뢰계수 .87보다 높은 것이었다. 한국어 도구의 구성타당도를 조사하기 위해 요인 분석을 실시한 결과 본 도구는 친밀한 주변인, 사회적 주변인, 및 소속감 등의 3개 요인을 가지는 것으로 분석되었다. 구성타당도에 대한 이러한 결과는 미국의 대학생들을 대상으로 한 Austin(1983)의 연구결과와 일치하는 것이었다. 각 요인에 대한 신뢰계수는 .88(제 1요인), .90(제 2요인), .84(제 3요인)로 나타났다. 전반적으로 대상자들은 한국어 도구에 대해 이해를 잘 하는 편이었으나 부정문인 4번 항목, '나는 혼자라고 느끼지 않는다'에 대해서는 응답을 할 때는 다른 항목에 비해 시간이 더 소요되거나 연구보조자에 의한 반복 설명이 필요하기도 하였다.

• The Journal of the Korea Contents Association
• /
• v.20 no.8
• /
• pp.350-358
• /
• 2020

The objective of this empirical study is to examine the mediating and moderating effects of perceived risks on the relationship between knowledge, feelings and acceptance intention towards AI. Subjects in their teens to forties were surveyed and the final sample comprised 1,969 subjects. Data were analyzed using Mediation using Multiple Regression and Moderated Multiple Regression. Results showed that people's knowledge and feelings towards AI affected their acceptance intention of AI. Results also showed that the perceived risks of AI partially mediated and moderated the relationship between feelings and acceptance intention towards AI and moderated but not mediated the relationship between knowledge and acceptance intention towards AI. Overall, these results suggest that people's perceived risks of AI are associated more strongly with their feelings towards AI than their knowledge towards AI. Implications and directions for future research were discussed in relation to increasing general population's acceptance intention towards AI.

• Information Systems Review
• /
• v.22 no.4
• /
• pp.163-184
• /
• 2020

The information on SNS, which is created by the forms of postings and comments, is being magnified and redistributed to news media expanding its impacts on real words. This amplifying effects of SNS comments have been increasingly discussed but there still lacks the answers for which dimensions of information is magnified, and what affects the direction and the degree of the amplification. This study, therefore, explores the detailed dimensions that are magnified by SNS comments and how SNS posting structure and social network characteristics affect them by using sentiment analysis. By analyzing 2,378 Facebook postings and news articles and their 26,312 SNS and 74,730 news site comments, this study shows that SNS comments magnify the sentiments of the posting articles they are attached to. In comparison to news site comments, SNS comments magnify the cognitive and social dimensions more than the news site comments. In the affective dimension, they tend to magnify only the positive emotion more than news site comments. Also, the findings reveal that whether the article in the posting is written by the posting owner affects the degree of amplification when the comments are remained positive or switched positive, while the opposite determines the amplification when comments remain negatively, suggesting that the user relationship in social network is the important factor that affects the direction and the degree of the information amplification in SNS.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.5
• /
• pp.15-30
• /
• 2023

Recently, generative models based on the Transformer architecture, such as ChatGPT, have been gaining significant attention. The Transformer architecture has been applied to various neural network models, including Google's BERT(Bidirectional Encoder Representations from Transformers) sentence generation model. In this paper, a method is proposed to create a text binary classification model for determining whether a comment on Korean movie review is positive or negative. To accomplish this, a pre-trained multilingual BERT sentence generation model is fine-tuned and transfer learned using a new Korean training dataset. To achieve this, a pre-trained BERT-Base model for multilingual sentence generation with 104 languages, 12 layers, 768 hidden, 12 attention heads, and 110M parameters is used. To change the pre-trained BERT-Base model into a text classification model, the input and output layers were fine-tuned, resulting in the creation of a new model with 178 million parameters. Using the fine-tuned model, with a maximum word count of 128, a batch size of 16, and 5 epochs, transfer learning is conducted with 10,000 training data and 5,000 testing data. A text sentiment binary classification model for Korean movie review with an accuracy of 0.9582, a loss of 0.1177, and an F1 score of 0.81 has been created. As a result of performing transfer learning with a dataset five times larger, a model with an accuracy of 0.9562, a loss of 0.1202, and an F1 score of 0.86 has been generated.

• Journal of Animal Science and Technology
• /
• v.45 no.2
• /
• pp.191-198
• /
• 2003

This study was carried out to investigate a usefulness of the microsatellite DNA markers for individual identification and parentage verification in three dog breeds. A total of 59 random dog (31 Chiwawa, 20 Poongsan, 8 Labrador Retriever) samples were genotyped by using 14 markers (Chiwawa dog), 16 markers (Poongsan dog), and 12 markers (Labrador Retriever dog) among the 17 international standard markers (PEZ1, 3, 5, 6, 8, 10, 11, 12, 13, 15, 16, 17, 20, 21, FHC2010, FHC2054 and FHC2079), respectively. The number of alleles per locus varied from 4 to 14 with a mean value of 6.07 in Chiwawa dog, 2 to 9 with a mean of 4.75 in Poongsan dog, and 3 to 5 with a mean of 4.00 in Labrador Retriever dog. Observed heterozygosity was ranged 0.419${\sim}$0.968 (mean 0.755), 0.300${\sim}$0.950 (mean 0.597) and 0.125${\sim}$0.750 (mean 0.604), and expected heterozygosity was ranged 0.432${\sim}$0.883 (mean 0.711), 0.262${\sim}$0.817 (mean 0.559) and 0.425${\sim}$0.808 (mean 0.660) in these three dog breeds. PIC value was ranged 0.397${\sim}$0.856 (mean 0.659), 0.222${\sim}$0.772 (mean 0.503) and 0.354${\sim}$0.717 (mean 0.563) in these three dog breeds. Of the 17 markers, PEZ1, PEZ3, PEZ6, PEZ10, PEZ12 loci, PEZ1, PEZ6, PEZ13 loci, and PEZ8, PEZ12 loci have relatively high PIC value (＞0.7) in Chiwawa dog, Poongsan dog and Labrador Retriever dog, respectively. The exclusion probability was ranged 0.240${\sim}$0.741, 0.111${\sim}$0.616, and 0.198${\sim}$0.529, and the combination of microsatellite loci was 0.9999, 0.9991, and 0.9968 in Chiwawa dog, Poongsan dog and Labrador Retriever dog, respectively. These results can give basic information for developing parentage verification and individual identification system in these three dog breeds.

• Journal of Korean Philosophical Society
• /
• v.143
• /
• pp.189-214
• /
• 2017

Disgust is caused by human being incompleteness and also its denial. Therefore the understanding of disgust is concerned with the understanding of humans. Thinking critically about disgust, we can solve the problems of disgust and reflect on our limitations. I will refer to Nussbaum's view which finds out in "projective disgust" an antipathy to human fault and a wish for completeness. This article is interested in the fact that disgust is not only a rejective feeling of a person but also a collective emotion which is connected with an antagonistic relationship and power exercise. Specifically, this article focuses on the point that disgust is mobilized to maintain a dominative relationship between humans. This is associated with the inner characteristics of disgust because disgust in itself contains a one-sided perspective, exclusiveness, hierarchy, and domination. This article aims to reveal a collusion between disgust and domination. For the purpose we will pay attention to two basic inclinations which are immanent in disgust; purity and exclusion, while relying on Nussbaum's view of disgust. In accordance with this analysis, this article will specificity treat disgust which can be summarized as an ideological function of emotion and its violent tendency. Then, we can ensure that disgust threats the equal and dignitary worth of human beings and hinders the diversity and rationality of a liberal society. In addition, it will be emphasized that disgust should not be used as an ideology which discriminates and suppresses a specific group in a society.

• Journal of Popular Narrative
• /
• v.26 no.2
• /
• pp.123-153
• /
• 2020

As the importance of documented fact has weakened in historical dramas, the combination with other genres has become prominent. By reviewing the way romance is dealt with in historical dramas, this research examines how the properties of historical events adopted by historical dramas are related to the motif of love, and how the narrative of love and romance contributes to the historical effects, with a focus on the television drama Mr. Sunshine. Mr. Sunshine is the first historical drama written by Kim Eun-sook, combining deliberately rearranged history with the writer's unique grammar of romance. The failed resistance movement of the righteous army in the drama is matched with the love that cannot be achieved based on self-negation. The drama, which deals with the tyranny of Japanese imperialism and the independence of Joseon, fictionalizes key characters and events, transforming the desire of love into the passion of patriotism. Romance in Mr. Sunshine serves as a catalyst for emphasizing the tragedy of historical events and reconstitutes cultural memories. In historical dramas, the fictional plot of romance leads viewers to reflect on human life in history that flows from the past to the future. How does an individual's inner feelings contribute to the historical representation? This research is significant as it is the first attempt to examine the relationship between historical drama and romance in various ways.

• The Journal of the Convergence on Culture Technology
• /
• v.4 no.3
• /
• pp.87-92
• /
• 2018

This paper aims to investigate political stances of news producers or news organizations by analyzing editorials from two Korean newspapers, Chosun Ilbo and the Hankyoreh, which are totally different in ideology, both qualitatively and quantitatively. In particular, the headlines from 16 editorials and 14 editorials from the two newspapers respectively published from May 28 to June 27 were analyzed in terms of political stances to a U.S-North Korea Summit. Moreover, two editorials published right after the U.S-North Summit were quantitatively analyzed within Martin and White (2005)'s framework. As a result, it was found that Chosun ilbo showed a negative stance to the summit by employing the 'feeling' factor within an attitude component most frequently, whereas the Hankyoreh was overwhelmingly positive toward the issue and the dialogue expansion factor within an engagement component is most frequently used to deliver its positive stance toward the issue.