• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.612-616
• /
• 2021

자연어 처리의 다양한 task 들을 잘 수행하기 위해서 텍스트 내에서 적절한 용어를 골라내는 것은 중요하다. 텍스트에서 적절한 용어들을 자동으로 추출하기 위해 다양한 모델들을 학습시켜 용어의 특성을 잘 반영하는 n 그램을 추출할 수 있다. 본 연구에서는 기존에 존재하는 신경망 모델들을 조합하여 자동 용어 추출 성능을 개선할 수 있는 방법들을 제시하고 각각의 결과들을 비교한다.

• Journal of Distribution Research
• /
• v.10 no.4
• /
• pp.89-111
• /
• 2005

The purpose of the study was to examine the role of cognitive costs in the relationship between service quality and store loyalty using structural equation modeling. The findings showed that cognitive costs playa mediating role between service quality and store loyalty. The findings also showed that physical environment indirectly affected store loyalty through cognitive cost. However, personal service had both direct effect and indirect effect on store loyalty. It was also appeared that there were differences in the models between customers and prospects.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.270-270
• /
• 2017

스마트워터그리드와 같은 첨단 정보통신기술을 활용한 물 관리 기술의 도입으로 수도운영사업에서도 누수와 같은 이상사건인지 목적의 효율적 빅 데이터 분석기법의 중요성이 증대되고 있다. 국내외적으로 누수인지를 위한 다양한 연구기법, 범위, 계측항목, 샘플링 주기 등이 제시된 바 있으나, 이상감지시스템(Event Detection System, EDS)은 대상지역 특정적 특성을 가지고 있어 범용적인 모델을 구축하는 데는 어려움이 있다. 본 연구에서는 소블럭 단위의 유량자료 분석을 통한 이상감지시스템의 적용가능여부를 판별하고 적합 모델구축자료 방안을 제시하는 K-EDS 모델을 개발하였다. 모델분석의 절차는 자료획득, 자료 전처리, 탐색적 자료해석, 그리고 각 기법 평가로 진행된다. 개발된 모델을 다양한 특성을 가지는 실제 지방상수도시스템에 적용하여 분석하였으며, 최종적으로 모델적용 가능성과 영향인자 등을 도출하였다. 개발된 모델은 소블럭별 현장계측자료 기반의 이상감지모델 적용 적합도 판별에 활용될 수 있으며, 향후 누수 인지 및 누수지속시간 감소를 위한 SW로 개발이 가능하다.

• Journal of Science Education
• /
• v.47 no.2
• /
• pp.192-210
• /
• 2023

This study was intended to develop the plant molecular biology experimental program using Rapid-cycling Brassica rapa (RcBr) based on the teaching steps and teaching methods of the cognitive apprenticeship model and to determine its application effects. In order to improve a subject's cognitive function and expertise on molecular biology experiments, two themes composed of a total 8 class sessions were selected: 'Identification of DFR gene in purple RcBr and non-purple RcBr' and 'Identification of RcBr's genetic polymorphism site using the DNA profiling method'. Research subjects were 18 pre-service teaching majors in biology education of H University in Chungbuk, Korea. The effectiveness of the developed program was verified by analyzing the enhancement of 'cognitive function' related to the use of molecular biology knowledge and technology, and the enhancement of 'domain-general metacognitive abilities.' The effect of the developed program was also determined by analyzing the task flow diagram provided. The developed program was effective in improving the cognitive functions of the pre-service teachers on the use of knowledge and technology of molecular biology experiments. It was especially effective to improve the higher cognitive function of pre-service teachers who did not have the previous experience. The developed program also showed a significant improvement in the task of metacognitive knowledge and in the planning, checking, and evaluation of metacognitive regulation, which are sub-elements of domain-general metacognitive abilities. It was found that the developed program's self-test activity could help the pre-service teachers to improve their metacognitive regulation. Therefore, this developed program turned out to be helpful for pre-service teachers to develop core competencies needed for molecular biology experimental classes. If the teaching and learning materials of the developed program could be reconstructed and applied to in-service teachers or high school students, it would be expected to improve their metacognitive abilities.

• Journal of Digital Convergence
• /
• v.18 no.1
• /
• pp.45-52
• /
• 2020

Online rumor creates psychological stress and image loss for victims. Prior studies related to online rumor did not consider the online environmental factor, despite the fact that online rumor occurs in the online space. Therefore, this study tried to investigate the influence of online characteristics on online rumor. This study considered perceived anonymity, lack of social presence, and perceived dissemination as online characteristics. We established and demonstrated a research model in which online characteristics affect online rumor behavior through attitude toward online rumor. This study obtained the sample of 201 social network users based on the survey and verified the research model using PLS tool. The results provided that perceived anonymity and perceived dissemination influenced online rumor behavior through attitude toward online rumor. On the other hand, lack of social presence was not significant. The findings of this study provide the fact that an individual's online rumor behavior can be caused by online characteristics. This study suggests that we pay attention to the role of perceived anonymity and perceived dissemination for online rumor behavior.

• The Journal of the Korea Contents Association
• /
• v.17 no.5
• /
• pp.440-448
• /
• 2017

This study aims to determine the factors affecting people's intention to use mobile delivery application services. The technology acceptance model (TAM) was employed and a survey was conducted on 300 delivery application users. As a result, personal innovativeness was found to have a positive effect on perceived ease of use and perceived playfulness. Self-efficacy turned out to have a positive influence on perceived ease of use and perceived playfulness. Next, perceived ease of use was found to positively affect both perceived usefulness and perceived playfulness, and perceived usefulness and playfulness had a positive effect on the intention to use the service as well. Lastly, personal innovativeness and self-efficacy were found to have a significant indirect influence on the intention to use the service. This study suggests users can be induced to voluntarily and continuously use the service, if their personal characteristics and perception of the service are considered in the future development of the service.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2013.11a
• /
• pp.34-35
• /
• 2013

기존 연구에서 무선 인지 네트워크의 성능 분석에 관한 연구는 많았지만 무선 인지 네트워크에 차량 통신을 접목시킨 무선 인지 차량 네트워크의 성능 분석에 관한 연구는 많지 않았다. 또한 최근 실질적인 채널환경을 고려하기 위해 불완전한 채널정보를 가잔 시스템에서의 성능 분석에 대한 연구가 많이 진행되고 있는데 무선 인지 차량 네트워크에서의 연구는 아직 진행되지 않았다. 본 논문에서는 불완전한 채널정보가 존재하는 인지 차량 네트워크의 성능 분석을 한다. 본 논문에서는 언더레이 (underlay) 무선 인지 네트워크를 가정하고 차량간 통신이 일어나는 2차 사용자 네트워크의 채널들을 double Rayleigh fading으로 모델링한다. 성능 분석의 지표로 불능 확률을 사용하고 컴퓨터 모의실험을 통해 분석된 불능확률 값이 상관계수가 변화함에 따라 어떠한 변화가 있는지 확인하였다.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.08a
• /
• pp.214-217
• /
• 1998

합성음에 사용할 한국어 억양 모델을 세우기 위한 기초적 연구로서 한국어 억양 곡선에 RFC 모델을 적용해 보았다. 억양 곡선의 구조는 피치 액센트와 억양구 경계 음조의 연속으로 되어 있는데, RFC 모델은 각각의 진폭과 지속기간을 가지는 상승 음조 요소와 하강 음조요소, 그리고 연결 요소로 이러한 억양 곡선의 모양을 모델링한다. 본 논문에서는 한국어 억양 곡선의 특징을 잘 반영하도록 RFC 모델의 구성요소를 수정하고, 청취 실험을 통해서 원래의 RFC 모델과 수정된 RFC 모델을 비교해 보았다. 실험 결과는 수정된 RFC 모델이 원래의 RFC 모델보다 13%정도 음조 표지 개수가 줄었음에도 불구하고 청각적으로 인지하는데 차이가 없는 것으로 나타났다.

• Science of Emotion and Sensibility
• /
• v.6 no.1
• /
• pp.17-26
• /
• 2003

With rapid expansion in e-retailing of apparel business, the virtual fitting model is used for comparing fit of the selected garments. The researchers experimented women's perceived body size for developing virtual fitting model. 50 women who were 20's in age were participated in the experiment They were graduate students with clothing & textile major. The results of this study showed that the subjects perceived themselves with fuller figure at waist or hips and with shorter figure at waist height than their actual body site. Many subjects believed that their body sizes were close to the average site. Especially, subjects whose abdomen girth of hip girth was much larger than average thought that they have relatively flat abdomen. It was concluded that not only the user's body measurements but also user's perception on their figure type are needed to be considered in the process of developing virtual fitting model for e-retailing of apparel. It was expected that the virtual fitting model of extraordinary size body would not be accepted by the female consumers even though they are extra large or small size person in reality.

• Korean Journal of Cognitive Science
• /
• v.2 no.1
• /
• pp.193-204
• /
• 1990

This paper discusses a reasoning model of information retrieval with a hierarchical thesaurus.The model computes the conceptual distance between a query and an object,both are indexed with weighted terms from a hierarchical thesaurus. The proposed model allows Boolean operators for user queries and edge weights for a hierarchical thesaurus. Experimental results have shown that the proposed model simulates, with surprising accuracy, people in the assessment of conceptual closeness between queries objects.