• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.11a
• /
• pp.185-188
• /
• 2010

온톨로지 간 이질성 문제를 해결하고 상호운용성을 향상시키기 위한 연구가 진행되어 왔으며, 최근 XMDR에 기반한 온톨로지 구축 방법이 제안되었으나 기존 연구와의 비교 평가가 부족하여 장점을 정확하게 보이지 못하였다. 따라서 이 논문에서는 XMDR 기반 온톨로지 구축 방법의 장점을 보다 명확하게 보이기 위해 정량적인 평가를 수행한다. 이를 위해 실제 온톨로지를 구축하고, 구축된 온톨로지는 온톨로지 참조 기반 온톨로지 구축 방법, 사전 참조 기반 온톨로지 구축 방법, 기존 방법론을 이용한 온톨로지 구축 방법을 평가 대상으로 하여 5가지 평가 지표로 분석된다. 평가 지표로는 구축된 온톨로지의 어휘 및 구조의 일관성 비교를 위하여 어휘 및 구조의 빈도수 평균과 엔트로피를 사용하고 구축 비용의 평가를 위하여 각 온톨로지의 구축 시간을 사용한다. 이러한 실험 및 평가의 결과로써, 온톨로지 참조 기반의 온톨로지 구축 방법은 다른 온톨로지 구축 방법들에 비해 온톨로지 어휘 및 구조가 일관적이고 효율적임을 보인다.

• Journal of Intelligence and Information Systems
• /
• v.16 no.1
• /
• pp.1-16
• /
• 2010

Several techniques and methods for ontology modularization have been proposed recently. However, there are few ontology evaluation frameworks to evaluate these techniques and methods. Most researches on ontology modularization have not been focused on ontology modularization evaluation but ontology modularization process itself. In this paper, we devise a novel ontology modularization evaluation framework to measure the quality of ontology modules, logical integrity during modularization process and modularization tools. Experiments were conducted to validate the proposed framework. Three representative modularization approaches SWOOP, Prompt, and PATO were chosen and used to partition or extract modules from an ontology. Then the proposed evaluation framework is applied to these modules. The experiment results indicate that the modularization framework works well. The proposed framework would help ontology engineers improve ontology module quality, anticipate and reduce future maintenance as well as help ontology users to choose ontology modules that best meet their requirements.

• Journal of the Korean Society for information Management
• /
• v.32 no.2
• /
• pp.193-219
• /
• 2015

The focus of traditional evaluations of ontologies is largely performance-based. A comparison of a new ontology with well-established ones, testing of ontologies in different applications, as well as any judgment of an ontology's appropriateness and relatedness to source data heavily rely on what results that ontology seems to manifest. This study, on the other hand, is an attempt to evaluate the quality of a particular ontology as manifested by its structure, representation, and interoperability. To that end, major categories of quality evaluations were first identified through an extensive survey of literature. Evaluation questions were formulated from these categories using the Delphi method and were validated by ontology experts. The entire process produced a set of 53 evaluation questions, which was then employed to test the quality of a newly-developed smartphone ontology.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.04a
• /
• pp.1017-1020
• /
• 2009

본 연구는 차세대 컴퓨터 패러다임의 지식베이스로 각광받는 온톨로지의 구축에 있어, 온톨로지가 실제 응용이나 서비스로 더욱 안정적으로 활용될 수 있는 온톨로지 평가를 위한 이론적 근거를 제시하는데 그 목적이 있다. 현재 많은 온톨로지가 구축되고 응용에 활용되기 위한 연구가 진행되나 이의 안정성과 유용성을 평가하기 위한 방법에 대한 연구는 부족한 상황이다. 따라서 본 연구는 소프트웨어공학적 측면에서 온톨로지의 소비자들이 만족할 수 있고 활용될 수 있는 온톨로지 구축을 위한 온톨로지를 평가하기 위한 서비스지향적 온톨로지 공학 기준과 온톨로지 평가 방안을 제시하고자 한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.04a
• /
• pp.664-667
• /
• 2014

기존 온톨로지들을 공유 및 재사용하기 위하여 온톨로지 정렬이 연구되고 있다. 기존 정렬 시스템은 온톨로지 데이터 양에 따라 매트릭스를 생성하고 과도한 계산을 통해 처리하여 대용량 데이터 집합에 대하여 공간적 및 계산적으로 부하를 발생하여 효율적이지 않다. 이를 해결하기 위하여 온톨로지 정렬을 휴리스틱 알고리즘을 적용하여 연구 진행하였다. 기존 휴리스틱 알고리즘은 계산이 간단하지만 조율해야 하는 파라미터가 많기에 특정 도메인에 최적 조합이 필요하며 만족한 성능을 얻지 못하였다. 이 논문에서는 Discrete Cuckoo Search(DCS) 기반 온톨로지 정렬 알고리즘을 제안한다. 제안한 알고리즘은 조율해야 하는 파라미터의 개수가 적고 Levy Flight 분포에 따라 탐색하여 계산이 간단하다. 제안된 알고리즘의 성능을 평가하기 위해 OAEI(Ontology Alignment Evaluation Initiative)에서 제공하는 벤치마크 데이터를 사용하여 정확률(Precision)과 재현율(Recall)을 구하고 기존 휴리스틱 정렬 알고리즘과 비교 평가하였다.

• Journal of Intelligence and Information Systems
• /
• v.14 no.4
• /
• pp.69-87
• /
• 2008

The Web is evolving toward the Semantic Web. Ontologies are considered as a crucial component of the Semantic Web since it is the backbone of knowledge representation for this Web. However, most of these ontologies are still built manually. Manual building of an ontology is time-consuming activity which requires many resources. Consequently, the need for automatic ontology extraction tools has been increased for the last decade, and many tools have been developed for this purpose. Yet, there is no comprehensive framework for evaluating such tools. In this paper, we proposed a set of criteria for evaluating ontology extraction tools and carried out an experiment on four popular ontology extraction tools (i.e., OntoLT, Text-To-Onto, TERMINAE, and OntoBuilder) using our proposed evaluation framework. The proposed framework can be applied as a useful benchmark when developers want to develop ontology extraction tools.

• Journal of KIISE:Software and Applications
• /
• v.34 no.12
• /
• pp.1021-1036
• /
• 2007

As the requirement of Semantic Web and knowledge management has been rising, ontology developments have been carried out actively. Ontology is now at the point that systematic developing standardization should be made up like the developing a large scale software. Yet, It has not been made to optimize and standardize ontology development methodologies. In this study, to compare ontology development methodologies, METHONTOLOGY, CommonKADS, OTK in foreign countries and EOE in Korea are selected. The evaluating method is to introduce CMM-I version 1.1 framework. Ontology development methodologies have been evaluated in process areas that introduce for CMM-I maturity level 2 and 3. The purpose of this study is to find matured process and weak process in ontology development methodology on the view of process maturity, and suggests the areas to be Improved in it. The result of this study can be applied as basic data to establish and improve ontology development methodology.

• 한국IT서비스학회:학술대회논문집
• /
• 2009.05a
• /
• pp.388-392
• /
• 2009

웹 관련 기술의 발전과 더불어 정보시스템의 개발에서 기계가 자동 처리할 수 있는 데이터의 기술 방법으로 온톨로지의 사용이 보편화되고 있다. 온톨로지는 특정 영역의 개념과 그들간의 관계를 단순 명료하게 기술한다. 지식 발견을 위한 도메인 온톨로지 구축은 도메인의 이해, 데이터의 이해, 테스크의 이해, 온톨로지 학습, 온톨로지 평가, 정제 등 다단계를 통해 완성되나 전문성이 요구된다. 본 논문에서는 학습 기반 도메인 온톨로지 구축방법을 제안하고 건강데이터를 위한 온톨로지 구축에서 응용하였다. 제안된 학습 기반 온톨로지 구축 방법은 건강데이터의 세부 영역별 개념과 관계를 밝히는데 유용하였다.

• Journal of the Korean Society for information Management
• /
• v.24 no.3
• /
• pp.343-362
• /
• 2007

The purpose of this study is to design and implement an ontology-based retrieval system for the electronic records of universities and to compare its performance with the existing keyword-based retrieval system. We used OntoStudio 1.4 for implementing an ontology-based retrieval system, and the test collection consisted of the following: (1) 5,099 electronic records of the 'personnel management notification' created by Korea Maritime University, (2) 20 topics (10 short-topics and 10 long-topics), and (3) the relevant assessments were conducted by the group of human experts. 10 university staff participated in the experiment of keyword-based searching and used the same test collection as used in the experiment of ontology-based searching. The ontology-based retrieval system outperformed to the keyword-based retrieval system in terms of Recall and Precision, and the same results showed in the test of the short-topics and long-topics comparison.

• The Journal of Society for e-Business Studies
• /
• v.15 no.2
• /
• pp.19-35
• /
• 2010

In response to an increased need, various methods for ontology modularization have been proposed. However, few studies have focused on evaluative methods for ontology modules. In this study, we devise novel metrics to measure ontology modularity. To evaluate the ontology modules, we introduce cohesion and coupling based on the theory of software metrics. A cohesion metric and two coupling metrics were used to measure cohesion and coupling for ontology modules. These metrics were also used to check consistency between the ontology modules and the original ontology. The new metrics support a more detailed relationship between classes in ontology modules. We validate the proposed metrics using the well known verification framework and perform the empirical experiments to complement previous investigations. This study offers ontology engineers valuable criteria with which to select and use ontology modules and modularization techniques.