• Knowledge Management Research
• /
• v.22 no.3
• /
• pp.91-106
• /
• 2021

With the recent rapid development of deep learning technology, the demand for analyzing huge text documents in the national R&D field from various perspectives is rapidly increasing. In particular, interest in the application of a BERT(Bidirectional Encoder Representations from Transformers) language model that has pre-trained a large corpus is growing. However, the terminology used frequently in highly specialized fields such as national R&D are often not sufficiently learned in basic BERT. This is pointed out as a limitation of understanding documents in specialized fields through BERT. Therefore, this study proposes a method to build an R&D KoBERT language model that transfers national R&D field knowledge to basic BERT using further pre-training. In addition, in order to evaluate the performance of the proposed model, we performed classification analysis on about 116,000 R&D reports in the health care and information and communication fields. Experimental results showed that our proposed model showed higher performance in terms of accuracy compared to the pure KoBERT model.

• Annual Conference on Human and Language Technology
• /
• 2003.10d
• /
• pp.95-100
• /
• 2003

본 논문에서는 현실 세계가 가지고 있는 지식이 어느 정도 체계적으로 정제되어 있는 국어사전, 백과사전 등을 중심으로, Hybrid Method를 이용한 통계(Statistics)기반 지식베이스와 어휘분류(Lexicon Classification)기반 지식베이스를 효율적으로 구축하여 질의응답시스템에 활용한다. 또한 특정한 문서를 보여주는 일반적인 질의응답시스템과는 달리, 이러한 지식베이스를 이용하여 사용자에게 정확한 개념어(정답어)를 습득하게끔 해주고, 사용자의 인지 체계 속에 어렴풋이 내포되어 있는 개념적 지식을 더욱더 표면적으로 확장해 나갈 수 있는 질의응답시스템을 구축하는 방안을 제시한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.321-323
• /
• 2000

국어 사전은 자연 언어 처리에서 필요로 하는 많은 정보를 구조적으로 포함하고 있으므로, 사전으로부터 다양한 언어 지식을 자동으로 획득할 수 있는 방법이 필요하다. 본 연구는 이러한 자동 지식 획득을 위한 기본적인 도구로서 국어 사전의 뜻풀이말 파서를 구현하는 것을 목적으로 한다. 이를 위해서 우선 국어 사전의 뜻풀이말을 대상으로 일정한 수준의 구문 부착 말뭉치를 구축하고, 이 말뭉치로부터 통계적인 방법에 기반하여 문법 규칙과 확률을 자동으로 추출한다. 본 연구는 이를 응용한 확률적 차트 파서를 구현하는 것이다. 그 결과 고려대 태거보다 11.61%의 정확률 향상을 보였는데, 이로써 구문 구조 정보가 품사 태깅에도 유용함을 알 수 있었다.

• Journal of Korea Entertainment Industry Association
• /
• v.14 no.6
• /
• pp.181-190
• /
• 2020

The study was conducted to examine the correlations between nursing students' knowledge, attitudes to advance medical directives and hospice perception. Data were collected using questionnaires from 181 students who were in 2rd and 4th year of the nursing college in D city, from May to Jun 2019. The collected data was analyzed using descriptive statistics, pearson's corelation coefficient and stepwise multiple regression with IBM SPSS 21.0 program. The study results showed that university students in nursing scored 3.52±.69, points for knowledge of advance directives, 3.10±.36 for attitude of advance directives, 3.14±.27 for hospice perception. The knowledge for advance directives was positively correlated with Attitude (r=.38, p <.001), and toward perception for hospice (r=0.28, p <.001). Based on the results of this study, it is necessary to develop a systematic education program in order to improve the awareness of the Perception of Hospice, Knowledge and Attitude toward Advance Directives.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.539-547
• /
• 2021

This study was conducted to identify the effects of knowledge of Advance Directives (AD) and the patient's perception of a peaceful death on their attitude toward the withdrawal of life-sustaining treatment and to provide basic data for the development of a nursing intervention program for activating self-determination in the withdrawal of life-sustaining support of patients. The subjects were 167 adult cancer patients who received outpatient or inpatient treatment, from September 15, 2019, to March 30, 2020. The data was analyzed by mean, standard deviation, t-test, ANOVA, Pearson's correlation coefficient, and multiple regression by using SPSS 21.0. From the results, it was observed that the knowledge of AD was 8.87±2.46 out of 12, perception of a peaceful death was 2.87±0.42 out of 4, and attitude toward withdrawal of life-sustaining treatment was 3.46±0.49 out of 5. There was a positive correlation between knowledge of AD, perception of a peaceful death, and their attitude toward withdrawal of life-sustaining treatment. The influencing variables were the knowledge of AD, perception of a peaceful death, discussion with family on withdrawal of life-sustaining treatment, and explanation power was 16.0% (F=10.355, p<.001). Therefore, it is necessary to develop a program that would improve the perception of a peaceful death, increase the knowledge of AD to improve the patients' attitude toward the withdrawal of life-sustaining treatment. An intervention to assist a discussion between the patients and their families in advance would also be useful.

• Korean Journal of Childcare and Education
• /
• v.9 no.2
• /
• pp.117-141
• /
• 2013

The purpose of this study is to investigate the young children's 'more means more' bias and knowledge change process regarding a lever phenomenon, especially the relationship between the weight of an object and the strength of force and between the length of a lever and the strength of force. Subjects, who were presented with the tasks, were eight young children 5 years of age. Major findings were as followings. First, most of the subjects have a 'more means more' bias about the relationship between the weight of an object and the strength of force and between the length of a lever and the strength of force regarding a lever phenomenon. This meant that young children have similar concepts about a lever phenomenon regardless of whether it is right or wrong physically. Second, young children tried to make sure of their knowledge during experiments. They chose the evidence which confirmed their knowledge. But they tried to change their knowledge, when the evidence presented did not correspond to their knowledge. These findings contribute to understanding young children's 'more means more' bias and knowledge change process about a lever phenomenon and can be used in preschool science education programs and curriculums.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.928-930
• /
• 2004

지금까지 기계번역과 교차언어 정보검색 등과 같은 자연언어응용에서 사용되는 번역지식을 자동으로 구축하는 연구가 활발히 진행되어 왔다. 번역지식을 자동으로 구축하는 연구는 대역사전에 등재되어 있지 않은 미등록어에 대한 대역정보를 문서에서 자동으로 획득하는 것을 목표로 한다. 최근에는 이러한 미등록어 중 음차표기 번역지식에 대한 연구가 활발히 진행되고 있다. 음차표기는 주로 영어 단어를 발음에 기반하여 비영어권의 언어로 표기하는 것을 의미한다. 음차표기된 단어들은 새로운 개념을 나타내는 신조어가 많기 때문에 사전에 등재되어 있지 않온 경우가 많다. 따라서 효과적인 번역지식 구축을 위해서는 이러한 음차표기 번역지식을 자동으로 획득하는 것은 매우 중요하다. 본 논문에서는 영-한 음차표기 대역쌍을 문서에서 자동으로 추출하는 알고리즘을 제안한다. 본 논문의 기법은 한국어 음차표기의 인식, 영-한 자동음차표기, 한국어 음차표기와 자동음차표기된 영어단어간의 음성적 유사도 비교를 통하여 음차표기 대역쌍을 추출한다. 본 논문의 기법은 약 93%의 정확률과 68%의 재현율을 나타내었다.

• The Journal of the Korea Contents Association
• /
• v.21 no.4
• /
• pp.688-698
• /
• 2021

The purpose of this study was to determine the effects of perception of good death and knowledge toward advance directive on attitude toward withdrawal of life-sustaining treatment among university hospital nurses. The subjects of this study were 207 university hospital nurses. Data were analyzed using the SPSS Statistics 24.0 version. The results showed that factors influencing attitude toward withdrawal of life-sustaining treatment were perception of good death (𝛽=-.32, p<.001), education experience for good death (𝛽=.15, p=.024), and knowledge toward advance directive (𝛽=.14, p=.036). They explained 14.2% of attitude toward withdrawal of life-sustaining treatment. Based on the results of this study, it suggests the development of education programs for good death and life-sustaining treatment to improve the perception of good death, knowledge toward advance directive of university hospital nurses.

• Journal of Digital Convergence
• /
• v.13 no.5
• /
• pp.187-194
• /
• 2015

Various linguistic knowledge resources are required in order that machine can understand diverse variation in natural languages. This paper aims to devise an automatic construction method of linguistic resources by reflecting characteristics of online contents toward continuous expansion. Especially we focused to build NE(Named-Entity) dictionary because the applicability of NEs is very high in linguistic analysis processes. Based on the investigation on Korean Wikipedia, we suggested an efficient construction method of NE dictionary using the syntactic patterns and structural features such as metadatas.

• 발명특허
• /
• v.34 no.7
• /
• pp.19-19
• /
• 2009