• Journal of Korean Society of Archives and Records Management
• /
• v.24 no.2
• /
• pp.89-112
• /
• 2024

Metadata is a crucial component of record management, playing a vital role in properly managing and understanding the record. In cases where automatic metadata assignment is not feasible, manual input by records professionals becomes necessary. This study aims to alleviate the challenges associated with manual entry by proposing a method that harnesses ChatGPT technology for extracting records management metadata elements. To employ ChatGPT technology, a Python program utilizing the LangChain library was developed. This program was designed to analyze PDF documents and extract metadata from records through questions, both with a locally installed instance of ChatGPT and the ChatGPT online service. Multiple PDF documents were subjected to this process to test the effectiveness of metadata extraction. The results revealed that while using LangChain with ChatGPT-3.5 turbo provided a secure environment, it exhibited some limitations in accurately retrieving metadata elements. Conversely, the ChatGPT-4 online service yielded relatively accurate results despite being unable to handle sensitive documents for security reasons. This exploration underscores the potential of utilizing ChatGPT technology to extract metadata in records management. With advancements in ChatGPT-related technologies, safer and more accurate results are expected to be achieved. Leveraging these advantages can significantly enhance the efficiency and productivity of tasks associated with managing records and metadata in archives.

• The KIPS Transactions:PartD
• /
• v.12D no.1 s.97
• /
• pp.135-150
• /
• 2005

In this paper, we propose an integrated information retrieval system for distributed multiple KRISTAL-2002 systems by using the metadata information. This system integrates current systems for different areas or systems for the same area with the different schemas so that the users can get the answers by once from the whole systems. The proposed system composes of the Source Server Manager(SSM) supporting the mapping between the integrated metadata database and source server, the Integrated Metadata Manager(ISM) for registering and managing the metadata and schema mapping, the Distributed Query Processor (DQP) for processing the user query into the source server query, the Distributed Data Set Integrated Manager(DDSIM) for transforming the total retrieval results by merging to the HTML format, and the integrated retrieval engine for managing the query results. It is assumed that the integrated metadata follows ISO/IEC 11179 metadata registration procedure with the metadata registry system which is a subsystem of the proposed system. There are two kinds of queries for users; the basic query and the detailed query. The users may select the databases or organizations for results by their own choices before giving the queries. The proposed system has been developed over KRISTAL-2002 systems with $Visual C^{++}\;and\;C^{++}-CGI$ and tested and verified with the six database systems.

• Journal of the Korean Society for information Management
• /
• v.27 no.3
• /
• pp.241-264
• /
• 2010

This study aimed to construct the Library Map Information System(LMIS) based on the Wiki theory of Web 2.0. We built this system because there was no collective source of information about every library in the world. Also, this system was developed to provide a library location information service by mashing-up with the Google Map. Through this study, the metadata applied to the newly constructed system was developed by using the Delphi method. A total of 13 experts including librarians of schools, public, academic, special, and national libraries as well as LIS faculty members and researchers, were commissioned as Delphi experts. Through three rounds of a Delphi survey analysis, the addition, modification, and deletion of the initial metadata elements was accomplished, and then the library contact/location information, library information, collection information, and event information was proposed. The metadata for LMIS was organized into four sectors and then 49 elements, each assigned to a sector.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.369-373
• /
• 2022

기존의 관계 추출 태스크에서의 많은 연구들은 사전학습 언어모델을 파인튜닝하여 뛰어난 성능을 달성해왔다. 하지만, 파인튜닝은 사전학습 시의 학습 기법과의 간극으로 인해 일반화 능력을 저해한다. 본 연구는 다운스트림 태스크를 사전학습의 Masked Language Modeling (MLM) 기법을 통해 해결하는 프롬프트 기반의 학습 기법을 활용하여, 특정 한국어 도메인에서의 관계 추출을 위한 프롬프트 기반 파인튜닝 방법론을 제안한다. 실험의 경우, 도메인의 특성이 뚜렷한 전통문화유산 말뭉치를 대상으로 실험을 진행하여 본 방법론의 도메인 적응력을 보이며, 메타 정보 즉, 개체 유형 및 관계 유형의 의미론적 정보를 일종의 지식 정보로 활용하여 프롬프트 기반 지식 주입의 효과성을 검증한다. 프롬프트에의 메타 정보의 주입과 함께 프롬프트 기반으로 파인튜닝된 모델은 오직 MLM 기법만을 이용하여 태스크를 수행하여 기존 파인튜닝 방법론 대비 파라미터 수가 적음에도, 성능 면에서 대부분 소폭 상승하는 경향을 보여줌으로써 그 효과성 및 효율성을 보인다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.688-690
• /
• 2002

인터넷과 웹의 팽창과 함께 가용 정보의 양이 폭발적으로 증가하고 있으나 이에 대응되는 효과적이고 효율적인 정보 검색능력의 지원이 없다면 이와 같은 방대한 정보들은 정보 이용자들에 있어 이용 가치가 없으며 이는 곧 정보 범람(information overflow)을 의미한다. 본 논문에서는 이에 대한 해결 방안으로써 사용자의 편이성과 정보검색 능력을 극대화할 수 있는 자연어 질의 문맥 구조 기반 개인형 메타 정보검색 엔진을 제안하고자 한다.본 방법론은 자연어 질의를 기본 입력 형태로 하여 자연어 질의의 문맥 구조(context structure) 및 기타 정보 평가 요소들을 이용하는 다척도(multi-criteria)의사 결정 기법 및 개인형 메타 정보 평가(information rating) 방법론으로 구성되어 있으며 이를 위한 시스템 설계를 제안한다.

• Smart Media Journal
• /
• v.10 no.2
• /
• pp.61-69
• /
• 2021

Wayfinding is a 'Process solving problem to find destination', and it is important to select spatial data for optimal way. Recently, due to the complexity of space and the expansion of the medium of the wayfinding service, it is necessary to the approach on the information design for them. Therefore, the purpose of this study is to propose a meta guideline on the information design for the design of orientation sign, which is an important cognitive clue in the wayfinding. It was conducted in 3 stages, First, a design process was proposed and design elements were derived for each step by literature research related to information and sign design, and analysis of manual for signage design. Second, a meta guideline for information organization and visualization in the three-stage design process was proposed by FGI and analysis. Third, the meta guideline was applied to the sign design on an area for user evaluation to inspect the applicability of the meta guideline. Through the user questionnaire, the possibility of applying the guideline for visualization of directions and spaces, information hierarchy according to spatial characteristics and information priority was identified. It is meaningful in that necessary element for signage design are systematized centering on the process and that it presents a macroscopic methodology according to spatial characteristics.

• Journal of the Korean Society for information Management
• /
• v.21 no.1
• /
• pp.231-251
• /
• 2004

This study aims to collect and manage in the step of creation metadata information to effectively manage and use official document which is a typical and normal records. To do it. data elements are extracted through analyzing structure of official document format. And we also select metadata elements reflecting creation background, publisher's intention, characteristic of official documents through evaluating and comparing extracted elements with data elements defined in ISAD rules. It would be draft data in constructing standardized metadata structure for records in Korea.

• Journal of the Korean Society for information Management
• /
• v.28 no.4
• /
• pp.161-182
• /
• 2011

Metadata and folksonomy are two main approaches in representing, organizing, and retrieving resources in the current information environment. Many researches have conducted studies to combine of metadata and folksonomy in order to utilize the strengths of both approaches. This research proposed an approach to utilize both metadata and folksonomy in representing resources by using microcontents. Microcontents in this research is a conceptual structure that reflects dynamic characteristics of folksonomy and the structure of metadata. By connecting folksonomy with metadata through this microcontents structure, both approaches can maximize their strengths and minimize their weaknesses in representing, organizing, and retrieving resources.

• Journal of the Korean Society for information Management
• /
• v.24 no.4
• /
• pp.191-221
• /
• 2007

The goal of this study was to examine the relationships between usability and the searching experience, and to uncover where an integrated cataloging and metadata system lay the usability problems of its primary users, catalogers. This study showed that the most important aspect of usability in Connexion lay in the experience of information retrieval system. Also, it showed there are seventeen usability problems to be improved in Connexion. Needless to say, it is most important to note that usability is not an exclusive goal of such an integrated cataloging and metadata system. Other goals such as quality of catalogs, and the reliability of its cataloging and metadata system are equally its concern. This study also suggested more testing on diverse cataloging systems and specific user groups, especially catalogers.

• Journal of Korea Society of Industrial Information Systems
• /
• v.13 no.4
• /
• pp.98-107
• /
• 2008

As the information and communications market goes more uncertain, foresight activities becomes more important. A number of foresight activities, such as trend analysis, have been used to predict customer needs. However previous studies tend to lack objectivity and systematization. In this study, we suggest a meta analysis methodology which combines both top-down and bottom-up approach in order to systematize the analysis process. Secondly, we applied this approach to ICT market to identify essential future technologies. Based on the result from the meta analysis, we have constructed the future technology roadmap.