• Annual Conference on Human and Language Technology
• /
• 2011.10a
• /
• pp.3-8
• /
• 2011

텍스트의 스토리를 자동으로 이해하기 위해 텍스트에서 기술된 사건(event)을 파악하고 이들을 조합하여 스토리가 어떻게 구성되어 있는지를 파악하는 연구들이 진행되어 왔다. 하지만 이는 스토리의 깊은 의미론적 이해를 요구하는 것 이외에도 텍스트마다 상황과 일어나는 사건들이 다양하기 때문에 언어 자원이 부족한 환경에서의 처리에는 한계가 있다. 이러한 문제는 사건들을 추상화 하여 단순하게 표현할 수 있다면 스토리 이해의 자연스러움을 저해하지 않고 해결 할 수 있다. 본 논문에서는 사건들의 추상화 과정을 위한 기초 연구로서 텍스트 속 등장인물이 행하거나 당하는 사건들을 추출하여 PMI기법을 통해 사건의 흐름을 파악하고 언어학적 단서를 참조하여 스토리 이해 과정에 누락될 수 있는 사건들을 추가하여 보완하였다. 이러한 접근을 통해 등장인물이 행할 수 있는 사건들을 재구성하여 단순화하는 방법을 제시한다.

• Annual Conference of KIPS
• /
• 2013.11a
• /
• pp.1121-1124
• /
• 2013

프로그램 실행 시 동화(텍스트)를 읽을 수 있게 하는 입력 파일을 받을 수 있게 만든다. 이렇게 받은 텍스트를 프로그램이 읽어 들여 알고리즘으로 대입한다. 알고리즘은 읽은 텍스트 중에 중요 단어를 뽑아내어 DB에 이에 알맞은 삽화를 요청하게 된다. 요청을 받은 데이터베이스는 다시 프로그램에 이미지를 보내주고, 프로그램은 이 삽화를 그 텍스트에 출력시키게 된다. 이 과정이 한 페이지를 만드는 과정이며, 원하지 않는 페이지는 건너 뛸 수 있게 한다. 이러한 작업을 반복하여 새로운 책 한권이 완성이 된다.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.64-68
• /
• 2020

텍스트 스타일 변환은 문장 내 컨텐츠는 유지하면서 문장의 스타일을 변경하는 것이다. 스타일의 정의가 모호하기 때문에 텍스트 스타일 변환에 대한 연구는 대부분 지도 학습으로 진행되어왔다. 본 논문에서는 병렬 데이터 구축이 되지 않은 데이터를 학습하기 위해 비병렬 데이터를 이용하여 스타일 변환을 시도한다. 트랜스포머 기반의 문장 생성기를 이용하여 문장을 생성하고, 해당 스타일을 분류하는 판별기로 이루어진 모델을 제안한다. 제안 모델을 통해, 감정 변환의 성능은 정확도(Accuracy) 56.9%, self-BLEU 0.393(긍정→부정), 0.366(부정→긍정), 유창성(fluency) 798.23(긍정→부정), 1381.05(부정→긍정)을 보였다. 본 연구는 비병렬 데이터에 대해 스타일 변환을 적용함으로써, 병렬 데이터가 없는 다양한 도메인에도 적용가능 할 것이다.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.171-189
• /
• 2022

Online reviews play an essential role in the consumer's purchasing decision-making process, and thus, providing helpful and reliable reviews is essential to consumers. Previous online review helpfulness prediction studies mainly predicted review helpfulness based on the consistency of text and rating information of online reviews. However, there is a limitation in that representation capacity or review text and rating interaction. We propose a CNN-RHP model that effectively learns the interaction between review text and rating information to improve the limitations of previous studies. Multi-channel CNNs were applied to extract the semantic representation of the review text. We also converted rating into independent high-dimensional embedding vectors representing the same dimension as the text vector. The consistency between the review text and the rating information is learned based on element-wise operations between the review text and the star rating vector. To evaluate the performance of the proposed CNN-RHP model in this study, we used online reviews collected from Amazom.com. Experimental results show that the CNN-RHP model indicates excellent performance compared to several benchmark models. The results of this study can provide practical implications when providing services related to review helpfulness on online e-commerce platforms.

• The Journal of the Korea Contents Association
• /
• v.21 no.11
• /
• pp.1-10
• /
• 2021

This study analyzes the use and meaning of consumer personality implication words in the review text of the Used Transaction Application . From of May 2021, the data were collected for the past six months by our Web crawler in Seoul and Gyeonggi Province, and a total of 1368 cases were collected first by random sampling, and finally 570 cases were preprocessed. The results are as follows. First, 48.2% of review texts were related to the personality of consumers even though it was a commercial platform of products. Second, the review text is mainly positive, which formed a text network structure based on the keyword 'gratitude'. Third, the review text, which implies consumer character, was divided into two groups: 'extrovert personality' and 'introvert personality' of consumers. And the individuality of the two groups worked together on the platform. In conclusion, we would like to suggest that consumer personality plays an important role in the platform transaction process, that consumer personality will play a role in the services of the platform in the future, and that consumer personality should be studied from various perspectives.

• The Journal of Information Technology
• /
• v.1 no.2
• /
• pp.55-63
• /
• 1998

본 논문은 웹 상에서 하이퍼텍스트 문서의 정보 검색에 있어서 검색에 요구되는 질의어에 따른 검색 결과가 주제에 따른 관련성을 측정하며, 하이퍼텍스트 문서가 링크되는 문서 상호간의 유사성에 대하여 정량화를 시도함으로써 검색 엔진의 성능분석을 제시한다.

• Archives of design research
• /
• v.14 no.4
• /
• pp.199-207
• /
• 2001

The public performance of drama has essence to convert text to visual images as stage and to represent, inform such visual images. Visual image is formed through a space as stage. Stage is a dish to fill the texts and is the mother's womb of visual image. Namely, visual image of drama equals space image. The purpose of this study is a trial to grasp the structure and system represented through interpreting. Especially, the concerns of this study are not semiology of letter imported in image but spacial image text importing just contents of text and the process and contents grasping the structure and meaning of dramatic space. Finally, this study proposed 'the system of space embodiment'from the semiotic point of view as interpretation methodology of actual memorial, symbolic space.

• Journal of the Korean Society for information Management
• /
• v.26 no.3
• /
• pp.111-129
• /
• 2009

This study examined the usability of text-based search and visual search of a large multidisciplinary library database to provide an empirical analysis of the acceptability of visual systems in the information retrieval environment. It also examined if there are differences in the usability assessment based on experimental order. The results indicated that the text-based search supported users' search behaviors more efficiently than the visual search. Also the text-based search was rated higher than the visual search in terms of user perceptions of four usability factors.

• The Journal of the Korea Contents Association
• /
• v.9 no.7
• /
• pp.149-158
• /
• 2009

Major purpose of this study is to explore effect of presentation modalities of Television and print on children's and adult's recall. So An experiment was conducted by comparing children's and adults' recall of information stories presented in three different modalities: "television moving Image1(auditory-visual redundancy)", "television moving Image2(auditory-visual redundancy)" and "print text". Results indicated that children remembered more infornation from the television moving Image than from print versions regardless of auditory-visual redundancy. But for the adults advantage of television was only found for information that had been accompanied by redundant pictures in television moving Image, providing support for the dual-coding hypothesis.

• Journal of the Society of Disaster Information
• /
• v.20 no.1
• /
• pp.13-19
• /
• 2024

Purpose: This study provides prevention methods by type of construction disaster using text mining techniques. Method: Based on the database that analyzed the cases of critical disasters in the domestic construction sector, preventive measures and causes are analyzed by text mining techniques, and the contents of the analysis are visually shown. Result: This visual data represents the measures for preventing critical disasters of each process according to the importance. Conclusion: It is believed that the results will be helpful in identifying factors to be considered in preparing preventive measures for serious accidents in construction.