• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.493-495
• /
• 2023

Deep learning methods and attention mechanisms have been incorporated to improve facial emotion recognition, which has recently attracted much attention. The fusion approaches have improved accuracy by combining various types of information. This research proposes a fusion network with self-attention and local attention mechanisms. It uses a multi-layer perceptron network. The network extracts distinguishing characteristics from facial images using pre-trained models on RAF-DB dataset. We outperform the other fusion methods on RAD-DB dataset with impressive results.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.4 no.2
• /
• pp.1-19
• /
• 2009

American Psychiatric Association defined hwa-byung as a unique mental illness of Korea, a kind of anger syndrome. This illness caused by the Korean culture is more often found on women than men, especially middle-aged women. The cause of hwa-byung mainly depends on the traditional family structure of Korea formed by Confucianism. Other causes that many middle-aged women suffer from hwa-byung are that their low education level originated in their wrong-recognized position in society and poverty in the past makes them not to be able to control their minds effectively(social psychological factor), and that they get stressed and depressed because of loss of their position in the family after their children leave home and the menopause which hurts their feminine beauty (physiological factor). So, this program aims at letting them to; 1) experience catharsis of the negative feelings which causes hwa-byung by expressing in psychodrama, 2) ease the symptoms of hwa-byung by relaxing their mind, 3) understand themselves through sharing their feelings and thinking at the group level, 4) acquire the opportunities to stabilize and expand their role in their real lives through the role playing.

• The Journal of the Convergence on Culture Technology
• /
• v.5 no.2
• /
• pp.117-122
• /
• 2019

This paper aims to investigate ideological preferences of news media outlets by looking at the news coverage of North Korea-Russia summit in April, 2019. The meeting has gained attention in South Korea, China, Japan and America in that the historical meeting will reflect the future direction for denuclearization on the Korean peninsular and the peace in the world. Given this, in particular, a special attention is paid to the editorials and headlines of news articles reported in two Korean quality newspapers, DongA Ilbo and Hankyoreh because both are quality newspapers, but are ideologically different. To achieve objectivity and fairness, the same issues dealt with during the summit were compared and analyzed within Martin and White (2005)'s Appraisal framework. As a result, it was found that in editorials of DongA Ilbo showed a negative stance to the summit by employing the 'attitude' factor, whereas the Hankyoreh was overwhelmingly positive toward the issue, also by employing the 'attitude' factor. The political stance is likely to be in line with those shown in the headlines of news articles from each newspaper. That is, it is clear that each news outlet shows its ideological stances to news consumers through linguistic expressions, in that both editorials and the headlines of news articles express their political preferences to the summit by means of linguistic appraisals.

• Journal of Practical Agriculture & Fisheries Research
• /
• v.21 no.2
• /
• pp.103-114
• /
• 2019

In this study we examined the preferences of college life factors for students at Korea National College of Agriculture and Fisheries(KNCAF). Analytical techniques of unstructured data used opinion mining and text mining techniques, and the results of text mining were visualized as word cloud. And those results were used for statistical analysis of the students' willingness to farm after graduation. The items of the favorable survey consisted of 10 items in 5 areas including university image, self-capacity, dormitory, education system, and future vision. After classifying the emotions of positive and negative in the collected questionnaire, a dictionary of positive and negative was created to evaluate the preference. The items of 'college image' at the time of university support, 'self after 10 years' after graduation, 'self-capacity' and 'present KNCAF' showed high positive emotion. On the other hand, positive emotion was low in the items of 'college dormitory', 'educational course', 'long-term field practice' and 'future of Korean agriculture'. In the cross-analysis of the difference in the will to farming according to gender, farming base, and entrance motivation, the will to farm according to gender and entrance motivation showed statistically significant results, but it was not significant in farming base. Also in binary logistic regression analysis on the will to farming, the statistically significant variable was found to be 'motivation for admission'

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.3
• /
• pp.207-218
• /
• 2020

Recently, due to the increase of online reviews and the development of analysis technology, the interest and demand for online review analysis continues to increase. However, previous studies have not considered the emotions contained in each vocabulary may differ from one reviewer to another. Therefore, this study first classifies the customer group according to the customer's grade, and presents the result of analyzing the difference by performing review analysis for each customer group. We found that the price factor had a significant influence on the evaluation of products for customers with high ratings. On the contrary, in the case of low-grade customers, the degree of correspondence between the contents introduced in the mall and the actual product significantly influenced the evaluation of the product. We expect that the proposed methodology can be effectively used to establish differentiated marketing strategies by identifying factors that affect product evaluation by customer group.

• Journal of the Korean Society for information Management
• /
• v.19 no.3
• /
• pp.123-143
• /
• 2002

Developing good methods for image description and indexing is fundamental for successful image retrieval, regardless of the content of images. Researchers and practitioners in the field of image indexing have developed a variety of image indexing systems and methods with the consideration of information types delivered by images. Such efforts in developing image indexing systems and methods include Panofsky's levels of image indexing and indexing systems adopting different approaches such as thesauri-based approach, classification approach. description element-based approach, and categorization approach. This study investigated users' perception of the objectiveness of image indexing, especially the iconographical analysis of image information advocated by Panofsky. One of the best examples of subjectiveness and conditional-dependence of image information is emotion. As a result, this study dealt with visual emotional information. Experiments were conducted in two phases : one was to measure the degree of agreement or disagreement about the emotional content of pictures among forty-eight participants and the other was to examine the inter-rater consistency defined as the degree of users' agreement on indexing. The results showed that the experiment participants made fairly subjective interpretation when they were viewing pictures. It was also found that the subjective interpretation made by the participants resulted from the individual differences in terms of their educational or cultural background. The study results emphasize the importance of developing new ways of indexing and/or searching for images, which can alleviate the limitations of access to images due to the subjective interpretation made by different users.

• Smart Media Journal
• /
• v.10 no.1
• /
• pp.16-24
• /
• 2021

The exiting word embedding methodology such as word2vec represents words, which only occur in the raw training corpus, as a fixed-length vector into a continuous vector space, so when mapping the words incorporated in the raw training corpus into a fixed-length vector in morphologically rich language, out-of-vocabulary (OOV) problem often happens. Even for sentence embedding, when representing the meaning of a sentence as a fixed-length vector by synthesizing word vectors constituting a sentence, OOV words make it challenging to meaningfully represent a sentence into a fixed-length vector. In particular, since the agglutinative language, the Korean has a morphological characteristic to integrate lexical morpheme and grammatical morpheme, handling OOV words is an important factor in improving performance. In this paper, we propose parallel Tri-LSTM sentence embedding that is robust to the OOV problem by extending utilizing the morphological information of words into sentence-level. As a result of the sentiment analysis task with corpus in Korean, we empirically found that the character unit is better than the morpheme unit as an embedding unit for Korean sentence embedding. We achieved 86.17% accuracy on the sentiment analysis task with the parallel bidirectional Tri-LSTM sentence encoder.

• Korean Journal of Environment and Ecology
• /
• v.15 no.2
• /
• pp.99-102
• /
• 2001

본 연구는 번식기와 비번식기에 있어서 까치의 행동변화를 파악하기 위해 강원도 춘천시 동면 감정리에 위치한 강원도 산림개발연구원 시험림 및 부근지역에서 2000년 3월부터 2001년 2월까지의 기간동안 실시되었다. 조사지 내 전체 면척은 5.6$\textrm{km}^2$였으며, 전체 지역에서 까치의 둥지는 모두 49개가 발견되었다. 이중 2000년 봄철의 번식기동안에는 24개의 둥지에서 까치가 번식을 한 것으로 나타났다. 까치의 번식밀도는 4.3쌍/$\textrm{km}^2$였으며, 둥지의 평균거리는 평균 357.5m 인 것으로 나타났다. 까치에 부착한 전파발신기에서 발신되는 전파를 추적하여 번식기와 비번식기에 둥지로부터 이동한 거리는 큰 차이를 보였다.(t-test, t= 2.98, P<0.01). 번식기에는 평균 133.5m의 거리를 둥지로부터 이동하였으며, 최대 184m 이동한 것으로 나타났다. 그 결과 세력권의 크기는 약 5.6ha인 것으로 추정되었다. 또한 비번식이에 있어서는 둥지로부터의 이동거리 평균 384.2m 였으며, 최대 457m 까지 이동하였다. 6월에서 8월 까지는 보통 2~3개체 정도가 무리를 형성하였고, 12월에 가장 많은 12개체 정도가 하나의 무리를 이루었다. 번식기에 암컷과 수컷을 행동은 매우 다른 것으로 나타났으나(t=2.89, P<0.05)비번식기에는 큰 차이가 없었다(t=1.03, P>0.05).

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.55-60
• /
• 2021

최근 한국어에 대한 자연어 처리 연구는 딥러닝 기반의 자연어 이해 모델을 중심으로 각 모델의 성능에 대한 비교 분석과 평가가 활발하게 이루어지고 있다. 그러나 한국어 생성 모델에 대해서도 자연어 이해 영역의 하위 과제(e.g. 감정 분류, 문장 유사도 측정 등)에 대한 수행 능력만을 정량적으로 평가하여, 생성 모델의 한국어 문장 구성 능력이나 상식 추론 과정을 충분히 평가하지 못하고 있다. 또한 대부분의 생성 모델은 여전히 간단하고 일반적인 상식에 부합하는 자연스러운 문장을 생성하는 것에도 큰 어려움을 겪고 있기에 이를 해결하기 위한 개선 연구가 필요한 상황이다. 따라서 본 논문은 이러한 문제를 해결하기 위해 한국어 생성 모델이 일반 상식 추론 능력을 바탕으로 문장을 생성하도록 KommonGen 데이터셋을 제안한다. 그리고 KommonGen을 통해 한국어 생성 모델의 성능을 정량적으로 비교 분석할 수 있도록 평가 기준을 구성하고, 한국어 기반 자연어 생성 모델의 개선 방향을 제시하고자 한다.

• Journal of the Korean Society for Library and Information Science
• /
• v.44 no.4
• /
• pp.203-230
• /
• 2010

This study analyzed the development and growth of folksonomy by examining tags associated with 40 bestsellers on LibraryThing.com in 6-month intervals. It was found that tag values do not decrease but grow in terms of quantity and quality. Accordingly, we examined the major significances of the tags and their potential utilization as an expression of subjects. Our findings were as follows. First, the motivations for tagging can be categorized into personal information for search purposes, self-fulfillment such as sense of achievement, display of emotion and sharing of one's experience with others, or an altruistic objective that emphasizes sociality with a desire that one's actions might provide social benefits. According to our analysis, 74.12% of tags had a social motivation. Second, the total number of tags and the frequency of usage increased with time. Third, the categories that showed a high increase in tag usage were dates of publication and reading, key words, main characters, and book reviews. Tags related to subjects had the highest ratio. Fourth, among Library of Congress Subject Headings (LCSH), multiple genres, key words and main characters were assigned to books, and specific key words and other properties were added as time progressed. There was also a slight increase in the number of tags consistent with LCSH. Fifth, we found that key tags could serve as a compilation of terms that reflects the knowledge base of the corresponding era. Thus, folksonomy should be continuously monitored for its quantitative and qualitative development of the tags to make improvements on its formative disadvantages, and identify internal semantic significance, be actively utilized in conjunction with taxonomy as a flexible compilation of terms that incorporate the history of a specific era.