• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.05a
• /
• pp.97-99
• /
• 2011

As interoperability between systems in distributed environment has been important, it has been possible for various organizations to share resources and exchange relevant information. However, semantic heterogeneity between the systems and organizations causes the problem of making their interoperability impossible. Thereby, in this paper, we propose an ontology matching-based knowledge management system which can automatically discover semantic correspondences between ontologies. Moreover, even though there have been many existing ontology matchers, it is still difficult to directly apply them to the proposed system. To deal with the problems, we want to discover matching patterns (MP) which they discover from two given ontologies.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.6
• /
• pp.721-729
• /
• 2006

Numbers of research on ontology construction and its application are being done for knowledge and information processing using computers. But, the current ontology development methods and ontology construction tools are using in restricted field on propose. Therefore, proper ontology development processes and ontology construction tools on ontology characteristic are needed. In this paper, we propose ontology construction process(OntoProcess) that non-experts in specific field are able to construct ontology through conceptualization of knowledge and formalization of concepts from language resource. Beside, some problems may be occurred while numbers of people are working together to construct ontology: i)duplicated concept definition in conceptualization process of knowledge and ii)decreasing efficiency of ontology construction by short understanding about formal language and tool operation in formalization process. To solve the problems, we propose an ontology construction process for multiple developers (OntoProcess) using meta ontology. We develop an ontology construction system(OntoCS) based on proposed processes, and we show the efficiency of proposed processes and system from ontology construction experiment.

• Journal of the Korean Society for information Management
• /
• v.25 no.3
• /
• pp.209-230
• /
• 2008

This study is based on the practical use for distributed resources considering growing network rapidly. The focal point of this study will be argued on semantic interoperability for sharing of resources, not be emphasized the technical issues of network. The aim of this article is developing the model of Collection-Level Description(CLD) for sharing of resources. The present article consists of a definition of collection in relation to the scope, objectives, and agents of the collection and an analysis of researches about CLD strengths and standards. Lastly, it was intended to construct the model focused on relation which was needed to be strengthened the existing CLD's function, thus, this study attempted to use the concept of ontology. The model of CLD based on ontology suggested the description could represent new relations inferred between classes and properties. Distinguishing class and property, furthermore, this study suggested properties were separated the characteristic of class and the relation with classes.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.3
• /
• pp.65-75
• /
• 2008

Grid Computing has enabled enormous amount of computational jobs by connecting distributed computing resources. This technology has developed and widely used in various fields. Previous researches usually focused on how to efficiently manage and use the grid resources. However, there was not enough tries to understand and manage information of application softwares in a well-defined structure. Therefore users in application domain need to how about grid deeply to identify and describe the resource requirements matching for each jobs. We introduce a semantic grid management system based on application ontology to overcome this problem. We design and implement the ontology to store various information of the applications. With the ontology, this system can infer the resource requirements from input parameters and input data of the application software and automatically assign appropriate resources by matching the requirement. Also it can transform the information to other forms which grid middlewares can handle. We apply the system to construct an analysis environment of bioinformatics and compare it with other grid systems to explain usefulness of the system.

• Journal of Korea Multimedia Society
• /
• v.13 no.9
• /
• pp.1263-1271
• /
• 2010

Ontologies, the important implementation tools for semantic web, are widely used in various areas such as search, reasoning, and knowledge representation. Developing well-defined ontologies, however, requires a lot of resources in terms of time and materials. There have been efforts to construct ontologies automatically to overcome these problems. In this paper, ontologies are automatically constructed from the natural languages sentences directly. To do this, the analysis of morphemes and a sentence structure is performed at first. then, the program finds predicates inside the sentence and the predicates are transformed to the corresponding ontology predicates. For matching the corresponding ontology predicate from a predicate in the sentence, we develop the "predicate ontology". An experimental comparison between human ontology engineer and the program shows that the proposed system outperforms the human engineer in an accuracy.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.05a
• /
• pp.183-186
• /
• 2006

독립적으로 구축되고 운영되어온 정보 자원의 공유와 통합된 서비스를 위해서 공통 온톨로지를 구축하고 활용하는 방법에 대한 연구가 활발히 진행되고 있지만, 온톨로지의 구축에 소요되는 많은 시간과 높은 비용은 온톨로지를 이용한 정보 통합의 있어 큰 장애가 된다. 본 논문에서는 UML 로 작성된 시스템 모델로부터 온톨로지 구축에 필요한 지식을 유도하여 초기 단계 도메인 온톨로지를 구축하는 방법을 제시하였다. 특히, UML 모델을 온톨로지 언어인 OWL로 변환할 수 있는 구체적인 방향을 제시함으로써, 향후 자동화된 도구 개발의 이론적 기반을 될 것으로 기대한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.706-708
• /
• 2005

시맨틱 웹의 등장으로 시맨틱 검색에 대한 관심이 높아졌다. 이에 본 논문에서는 시맨틱 웹 환경의 자원들을 이용하여 시맨틱 검색을 수행하는 검색방법을 제안한다. 기존의 SPARQL(Simple Protocol and RDF Query Language) 시맨틱 질의언어는 추론의 기능 없이 RDF(Resource Description Framework)에서 제공하는 SPO(subiect, predicate, object) 형태의 트리플 패턴 매치만을 제공한다. 본 논문의 시스템은 기존의 SPARQL질의 시스템에 시맨틱 추론기능을 추가하여 검색 결과에 효율성을 증가 시키는 것을 목적으로 한다. 본 논문에서의 시스템은 다음과 같은 특징을 갖는다. 첫째, 시맨틱 웹 환경의 온톨로지 구축을 위해서 W3C에서 온톨로지 언어로 표준화된 OWL(Web Ontology Language)를 사용하여 검색 환경을 구축한다. 둘째, 온톨로지와 메타데이터를 추론하여 시맨틱 검색을 유도하는 OWL추론기를 사용한다. 시맨틱 추론은 온롤로지의 공리(Axiom)을 충분히 활용하는 온톨로지기반 시맨틱 추론과 검색 도메인에 맞는 규칙을 활용하는 사용자 컨텍스트 기반의 시맨틱 추론으로 이루어진다. 셋째, 다양한 시맨틱 검색을 위해 W3C에서 제안되어 차세대 시맨틱 검색 질의언어로 연구중인 SPARQL을 사용한다. 이와 같은 특징은 시맨틱 검색 시스템이 시맨틱 웹 환경의 자원을 충분히 활용하는 결과를 가져온다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10a
• /
• pp.121-123
• /
• 2003

본 연구에서는 현재 웹 환경에서의 단순한 string matching 검색에 대한 한계점을 해결하기 위해서 온톨로지 기반의 자원검색 방안을 논한다. 테스트 베드로서 제주도의 숙박, 관광정보에 따른 온톨로지를 DAML+OIL언어로 생성하고 Jena에서 지원하는 API를 이용하여 사용자가 원하는 정보검색을 수행할 수 있는 테스트 베드 구축 방안도 제시한다.

• Annual Conference on Human and Language Technology
• /
• 2009.10a
• /
• pp.126-129
• /
• 2009

YKanto (Ontology Construction System of Yeungnam University & KAIST)[1,2]는 웹 기반 온톨로지 구축 시스템으로써 다수의 온톨로지 구축자가 신문, 사전, 코퍼스 등의 지식 자원으로부터 대용량의 온톨로지를 구축하고 관리할 수 있는 환경을 제공한다. 본 논문에서는 YKanto에서 개발 중인 온톨로지를 2차원 그래프로 시각화 할 수 있는 온톨로지 시각화 도구(Visualization Tool for Ontology)를 설계하고 개발한다. 개발한 온톨로지 시각화 도구는 기존의 개념 및 관계 중심의 시각화 도구와 달리 사건 기반의 온톨로지(event-based ontology)의 가독성을 높이기 위한 사건(event) 중심의 시각화 도구이다. 또한, Java Applet으로 구현하며, YKanto과 연동시킴으로써 개발 중인 온톨로지를 웹 상에서 실시간으로 확인할 수 있도록 한다.

• Journal of KIISE:Databases
• /
• v.36 no.5
• /
• pp.386-398
• /
• 2009

This paper proposes an ontology population model based on ISO/IEC 11179. Much research has recently been done on harmonizing Web 2.0 and the Semantic Web, and the harmonization is defined as Web 3.0. The most important issues for realizing Web 3.0 include defining ontology schemas and populating instances for ontologies. To resolve the issue, Web ontology schemas should be precisely defined, and a method for populating Web ontology from Web resources should be developed. This paper proposes a Web ontology population model based on ISO/IEC 11179 - Metadata Registry (MDR), which is the international standard, developed to manage and use common standard concepts.