• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.5
• /
• pp.31-39
• /
• 2023

The evolution of Natural Language Processing (NLP) and the rise of large language models (LLM) like ChatGPT have paved the way for specialized question-answering (QA) systems tailored to specific domains. This study outlines a system harnessing the power of LLM in conjunction with document search algorithms to interpret and address user inquiries using documents from the Korea Atomic Energy Research Institute (KAERI). Initially, the system refines multiple documents for optimized search and analysis, breaking the content into managable paragraphs suitable for the language model's processing. Each paragraph's content is converted into a vector via an embedding model and archived in a database. Upon receiving a user query, the system matches the extracted vectors from the question with the stored vectors, pinpointing the most pertinent content. The chosen paragraphs, combined with the user's query, are then processed by the language generation model to formulate a response. Tests encompassing a spectrum of questions verified the system's proficiency in discerning question intent, understanding diverse documents, and delivering rapid and precise answers.

• Proceedings of the KWS Conference
• /
• 2002.10a
• /
• pp.15-22
• /
• 2002

The modern welding technology hue more then 100 years history, and during that period, the welding has been playing an important role in the development of the manufacturing processes/systems as a key technology. In the time we enter the 21st century, it is question how to develop the innovative and attractive welding technology, in particular by the younger generation. The present report discussed the future trend of welding engineering mainly in Japan after reviewing the previous growth of welding engineering.

• Proceedings of the KIEE Conference
• /
• 2003.07b
• /
• pp.1389-1391
• /
• 2003

The development of wind turbine power generation has grown during the past ten years. An important question, when installing wind turbines with the generator connected directly to the grid, is holt much the voltage quality will be affected by the uneven power production and by the connection of the wind turbine to the grid. This paper presents the voltage fluctuation of grid-connected WTG(wind turbine generators) by MATLAB/Simulink.

• Journal of The Korean Association For Science Education
• /
• v.23 no.3
• /
• pp.215-228
• /
• 2003

The present study is to analyze the arguments about the generation of declarative scientific-knowledge in the philosophy of science and invent a structured model of the process of scientific-knowledge generation with the types of the generated scientific-knowledge. The invented model shows that scientific-knowledge generation is a distinctive process with the processes of inductive, abductive, and deductive thinking. Furthermore, inductive process is included with observation, which is consisted of simple observation and operative observation, and rule-discovery which is involved with the processes of commonness discovery, classification, pattern discovery, and hierarchical relationship. Also, abductive process has two components. One component generates question and second component generates hypothesis in which the process consists of representing question situation, identifying experienced situation, identifying causal explicans, and generating hypothetical explicans. Finally, deductive process is involved with logical inventing test method and evaluation criteria, concrete inventing test method and evaluation criteria, evaluating hypothesis, and making conclusion.

• Genomics & Informatics
• /
• v.11 no.4
• /
• pp.191-199
• /
• 2013

High-throughput next-generation sequencing (NGS) technology produces a tremendous amount of raw sequence data. The challenges for researchers are to process the raw data, to map the sequences to genome, to discover variants that are different from the reference genome, and to prioritize/rank the variants for the question of interest. The recent development of many computational algorithms and programs has vastly improved the ability to translate sequence data into valuable information for disease gene identification. However, the NGS data analysis is complex and could be overwhelming for researchers who are not familiar with the process. Here, we outline the analysis pipeline and describe some of the most commonly used principles and tools for analyzing NGS data for disease gene identification.

• Korean System Dynamics Review
• /
• v.12 no.4
• /
• pp.93-124
• /
• 2011

The introduction of competition in the generation of electricity has raised the fundamental question of whether markets provide the right incentives for the provision of the capacity needed to maintain system reliability. Capacity mechanisms are adopted around the world to guarantee appropriate level of investment in electricity generation capacity. In this study, we discuss these approaches and analyze the capacity pricing mechanisms from the adequacy perspective. We conclude that the design of capacity mechanism is very important to decrease electricity spot price and increase total electric capacity. Specifically, the constant of capacity pricing mechanism made a difference to the performance of electricity market. However, the slope of capacity price mechanism is better than the constant of that in improving performance of electricity market.

• Development and Reproduction
• /
• v.27 no.1
• /
• pp.47-56
• /
• 2023

Despite the commercialization of Next generation sequencing (NGS) gene testing, only a few studies have addressed the various ethical and legal problems associated with NGS testing in Korea Here, we reviewed the normative issues that emerged at each stage of the wet analysis and bioinformatics analysis of NGS gene testing. In particular, it was in mind to apply various international guidelines and the principles of bioethics to actual clinical practice. Considering the characteristics of NGS testing, wet analysis of additional testing can be justified if presumptive consent is recognized. Furthermore, the medical relationship between diseases needs to be established and it should be clear that the patient would have given consent if the patient had been aware of the correlation between genes. At the stage of bioinformatics analysis, the question of unsolicited findings arises. In case of unsolicited and relevant findings, according to American College of Medical Genetics and Genomics (ACMG), a recognized relationship between genes and diseases needs to be established. In case of unsolicited and not-relevant findings, it is almost impossible to determine whether knowing or not knowing the findings is more beneficial to the patient. However, it seems to be certain that the psychological harm an individual may suffer from such information is likely to be greater if the disease is severe and if there is no cure. The list of genes for which the ACMG guidelines impose reporting obligations is a good reference for judgment.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.6 no.6
• /
• pp.966-973
• /
• 2002

This Pager presents an object-oriented analysis and design for Remote Examination System on Internet. The UML is used for describing the process of analyzing and designing the system. The system is composed of the parts: Question Generation part that professors make questions, Question Answer part that students answer the questions. By object-oriented analysis and design, the system inherits the advantages of object-oriented methodology. All the results are retained on database on server-side. This system can be integrated with the Cyber-University System, and substituted for off-line paper-based tests or adopted to on-line poll applications on Internet.

• Annual Conference on Human and Language Technology
• /
• 2002.10e
• /
• pp.176-183
• /
• 2002

전자우편과 같이 일정한 질의 형식을 가지고 있는 긴 자연어 질의에 대해서 사용자 질의 단어에 가중치를 부과하는 방법과 질의에 대한 정답을 기존의 질의응답 집합에서 유사한 질의를 검색하여 그 정답을 사용자에게 제공하는 전자우편 질의응답 시스템을 제안한다. 사용자의 긴 자연어 질의가 주어지면 질의의 범주와 문장의 중요도 정보를 이용하여 질의에서 사용된 단어가 주제어로 쓰였을 확률을 계산하고, 계산된 확률에 기반하여 중요도를 할당하는 질의생성 모델을 제안한다. 또한 사용자 질의와 기존에 문의되어진 전자우편 질의의 유사도를 단어의 빈도를 고려한 어휘유사도, 한글 시소러스(Thesaurus)를 이용한 의미유사도와 본 논문에서 제안한 질의생성 모델을 이용한 주제 유사도를 이용하여 계산한다. 실험을 위하여 실세계에서 사용 중인 질의응답 집합을 이용하여 실험을 하였으며 각 유사도 계산 방법의 기여도를 비교 평가하고 제안한 질의생성모델이 성능향상에 미치는 영향을 평가하였다.

• Korean Journal of Acupuncture
• /
• v.32 no.2
• /
• pp.41-50
• /
• 2015

Objectives : In this study we investigated characteristics of eight extra meridians based on the chapter of Pyoung Gi Kyoung Pal Maek Byoung(PGKPMB) of the Pulse Classic(Maijing; PC) and the 27, 28, and $29^{th}$ question of the Classic of Difficult Issues(Nanjing; CDI). Methods : We searched out the contents of eight extra meridians in historical background, the principal and circulation of eight extra meridians, and the symptoms and pulse diagnosis of eight extra meridians on the PGKPMB of the PC and the 27, 28, and $29^{th}$ question of the CDI. Results and Conclusions : About development of eight extra meridians, content which was scattered in the CDI was supplemented and developed in the CDI. Also taking over the content in the CDI, the meaning of the eight extra meridians was expanded and supplemented in the PC. In the PC, the mechanism about symptom of eight extra meridians especially has been described. Also materialization of symptom which linked to pathogenic pulse of eight extra meridians had a great impact on future generation.