• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.8 no.1
• /
• pp.1-5
• /
• 2008

Generally the analytical tools of data mining have two learning types which are supervised and unsupervised learning algorithms. Classification and prediction are main analysis tools for supervised learning. In this paper, we perform a comparison study of classification algorithms in data mining. We make comparative studies between popular classification algorithms which are LDA, QDA, kernel method, K-nearest neighbor, naive Bayesian, SVM, and CART. Also, we use almost all classification data sets of UCI machine learning repository for our experiments. According to our results, we are able to select proper algorithms for given classification data sets.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2001.10a
• /
• pp.128-131
• /
• 2001

With the advent of the Internet, data centric computing applications, and e-business applications, virtually all network-stored data has become mission critical in nature. This increasing reliance on the access to enterprise data is challenging the limitations of traditional server-storage solutions. As a result, the ongoing need to add more storage, serve more users and back up more data has given rise to the concept of a SAN(Storage Area Network), “a network whose primary purpose is the transfer of data between computer systems and storage elements and among storage elements.” The purpose of this article is to help the managers and researchers make the understanding of a SAN as the possibility of new data repository in e-business era. This paper provides a thorough literature review on this new concept including its structure, its benefits and disadvantages, and its underlying issues.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.23 no.58
• /
• pp.1-16
• /
• 2000

To develop a generic ERP(Enterprise Resource Planning) system which can accommodate various types of manufacturing enterprises, object-oriented methods are commonly applied from analysis to implementation. The objective of OO-ERP (Object-Oriented ERP) systems is the reusability of business objects(components). In practice, one of the critical features for the reusable OO-ERP system would be the capability of interfacing with distributed, heterogeneous data repositories. Consequently, it is essential to provide data repository transparency in OO-ERP systems - business objects do not take care of the locations and types of data repositories. In this paper, we propose Data Object Layer(DOL) that supports such transparency. DOL is a horizontal component through which OO-ERP systems can be seamlessly connected with diverse data repositories.

• Korean Journal of Computational Design and Engineering
• /
• v.10 no.2
• /
• pp.97-106
• /
• 2005

W3C has developed the RDF standard for utilizing ontology in Web applications. This paper presents extracting, storing, and applying ontology on product data. The design and tooling information included in STEP-NC files is focused as an example. By analyzing the relationship among the product data, the RDF schema is designed first. Based on the schema ontology if extracted and stored in XML files. As an application of the stored ontology, we can reconfigure the sitemap of product data repositories. In this example, the users can select the view that he or she is interested in (e.g. product, tool, person). With such various views of an product data repository, the users can access the specific data more effectively.

• Tunnel and Underground Space
• /
• v.11 no.1
• /
• pp.42-58
• /
• 2001

In order to protect the environment from waste disposal activities, the prediction of the flux and flow paths of the contaminants from underground facilities should be assessed as accurately as possible. Especially, the prediction of the pathways and travel times of the nuclides from high level radioactive wastes in a deep repository to biosphere is one of the primary tasks for assessing the ultimate safety and performance of the repository. Since the contaminants are mainly transported with groundwater along the discontinuities developed within rock mass, the characteristics of groundwater flow through discontinuities is important for the prediction of contaminant fates as well as safety assessment of a repository. In this study, the actual fracture network could be effectively generated based on in situ data by separating geometric parameter and hydraulic parameter. The calculated anisotropic hydraulic conductivity was applied to a 3D porous medium model to calculate the path flow and travel time of the large studied area with the consideration of the complex topology in the area. Using the model, the pathways and travel times for groundwater were analyzed. From this study, it was concluded that the suggested techniques and procedures for predicting the pathways and travel times of groundwater from underground facilities to biosphere is acceptable and those can be applied to the safety assessment of a repository for radioactive wastes.

• Journal of the Korean Geotechnical Society
• /
• v.33 no.7
• /
• pp.55-64
• /
• 2017

A geological repository has been considered one of the most adequate options for the disposal of high-level radioactive waste. A geological repository will be constructed in a host rock at a depth of 500~1,000 meters below the ground surface. The geological repository system consists of a disposal canister with packed spent fuel, buffer material, backfill material, and intact rock. The buffer is very important to assure the disposal safety of high-level radioactive waste. It can restrain the release of radionuclide and protect the canister from the inflow of groundwater. High temperature in a disposal canister is released into the surrounding buffer material, and thus the thermal transfer behavior of the buffer material is very important to analyze the entire disposal safety. Therefore, this paper presents a thermal conductivity prediction model for the Kyungju compacted bentonite buffer material which is the only bentonite produced in Korea. Thermal conductivity of Kyungju bentonite was measured using a hot wire method according to various water contents and dry densities. With 39 data obtained by the hot wire method, a regression model to predict the thermal conductivity of Kyungju bentonite was suggested.

• Journal of the Korean Society for Library and Information Science
• /
• v.58 no.2
• /
• pp.151-175
• /
• 2024

This study analyzes the current status of research data management at NST-affiliated government-funded research institutes for the purpose of promoting the sharing and use of research data, and based on this, suggests methods for establishing a research data sharing and management system. The survey on the status of research data management was conducted twice in 2022 and 2023 for a total of 20 research institutes. In addition, difficulties and areas that need to be improved in the management and sharing of research data were identified, and based on this, methods for establishing a research data sharing and management system were proposed by dividing them into policy aspects, system aspects, and linkage system construction aspects. In order to establish a research data sharing system, it would be desirable to prepare a policy basis and present contents such as the definition of research data, scope of application, contents of management, utilization method, and leading institutes. In addition, for systematic and unified research data management, it would be recommended that each institute will establish and manage a repository and management system. By linking this with DataON, the national research data platform, and providing one-stop services, the accessibility and usability of data will be improved.

• Proceedings of the Korea Contents Association Conference
• /
• 2016.05a
• /
• pp.15-16
• /
• 2016

본 연구의 목적은 글로벌 리포지터리의 현황 정보를 수집하기 위한 모듈을 개발하고 이의 완전성을 검증하는데 있다. OpenDOAR와 ROAR 레지스트리에 등록되어 있는 리포지터리 정보를 수집하기 위해서 각각의 서비스에서 제공되는 API와 웹페이지에 대한 분석을 진행 하였다. 개발 모듈로는 대륙별 국가정보 수집모듈, 리포지터리 소프트웨어 정보 수집모듈, 리포지터리 주제 정보 수집모듈, 리포지터리 상세정보 수집모듈을 개발하였다. 수집모듈로 수집된 리포지터리 정보를 대상 정보원의 데이터와 비교하는 방법으로 모듈의 완전성을 검증하였다.

• The Journal of Information Technology and Database
• /
• v.7 no.1
• /
• pp.113-130
• /
• 2000

As software projects become larger and more complex, we need to take a more systematic approach to quality assurance. One plausible approach is the use of SQA (software quality assurance) methodology. Since this SQA methodology was not available, our study presents a SQA methodology. This methodology has a repository in which a set of quality assurance tasks with their related techniques and deliverables is defined and from which one can select only appropriate tasks based upon characteristics of project. This study further suggests a rule-based approach for supporting task selection process.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04a
• /
• pp.686-688
• /
• 2003

워크플로우 시스템의 핵심을 이루고 있는 엔진의 효율성을 극대화 시키기 위하여, 워크플로우 시스템 엔진 안에서 운용되는 데이터의 관리는 매우 중요하다. 본 논문에서는 워크를로우 시스템에서 운용되는 각 시스템 데이터의 특징을 고찰 한 후, 일반적으로 사용하는 데이터 베이스 시스템 호출을 통해 데이터를 관리하는 방법을 보완 할 수 있는 방법을 제시한다. 그 방법으로 각 시스템 데이터가 가지고 있는 특성에 맞추어 기존의 데이터 베이스 호출을 통한 방법에 메모리에 데이터를 로드시켜 공용적으로 사용하는 방법을 더해 시스템 데이터를 관리하는 방법을 기술한다.