• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.535-539
• /
• 2020

다양한 발화를 모델링하는 요구는 자연어 처리 분야에서 꾸준히 있었으며 단어, 구 또는 문장과 동등한 의미 콘텐츠를 자동으로 식별하고 생성하는 것은 자연어 처리의 중요한 부분이다. 본 논문에서는 포인터 생성 네트워크(Pointer Generate Nework)를 이용하여 패러프레이즈 생성 모델을 제안한다. 제안한 모델의 성능을 측정하기 위해 사람이 직접 구축한 유사 문장 코퍼스를 이용하였으며, 토큰 단위의 BLEU-4 0.250, ROUGE_L 0.455, CIDEr 2.190의 성능을 보였다. 하지만 입력 문장과 동일한 문장을 출력하는 문제점이 존재하여 빔서치(beam search)를 적용하여 입력 문장과 비교하여 생성 문장을 선택하는 방식을 적용하였다. 입력 문장과 동일한 문장을 제외한 문장으로 평가를 진행했으며, 토큰 단위의 BLEU-4 0.234, ROUGE_L 0.459, CIDEr 2.041의 성능을 보였으나, 패러프레이즈 생성 데이터 양이 크게 증가하였다. 본 연구는 문장 간의 의미적으로 동일한 정보를 정확하게 추출할 수 있게 됨으로써 정보 추출, 온톨로지 생성에 도움이 될 것이다. 또한 이러한 기법이 챗봇에서 사용자의 의도 탐지 및 MRC와 같은 자연어 처리의 여러 분야에 유용한 자원으로 사용될 것이다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2006.06a
• /
• pp.237-244
• /
• 2006

시맨틱 웹 관련연구가 증가함에 따라 하나의 관련분야로 규칙기반 시스템 동의 지능적인 웹 환경에 대한 기대 역시 커지고 있다. 하지만 규칙기반 시스템을 활용하기에는 아직도 규칙습득이 많은 제약이 되고 있다. 규칙습득은 웹으로부터 필요한 규칙을 습득하는 일련의 방법인데, 이러한 규칙을 습득하기 위해서는 규칙구성요소를 먼저 식별해야만 한다. 그러나 이러한 규칙을 식별하는 작업은 대부분 지식관리자의 수작업에 의해 이루어지고 있다. 본 연구의 목적은 웹으로부터 규칙구성요소 식별을 최대한 자동화하고 지식관리자의 수작업을 최소화함으로써 그 부담을 줄여 주는 데 있다. 이러한 방법으로는 온톨로지를 근간으로 하여 웹 페이지와의 문자열 비교, 이러한 비교의 한계를 극복하기 위한 확장등의 방법이 있다. 첫 번째 방법은 온툴로지 기반으로 규칙식별 할 웹 페이지와 비교를 통해 지식관리자의 규칙식별 과정을 최대한 자동화하여 주는 것이다. 여기서 만약 현재 규칙을 식별하고자 하는 웹 사이트와 유사한 시스템의 규칙들을 활용하여 일반화 된 온툴로지가 구축되었다면, 이 온톨로지를 기반으로 규칙을 식별하고자 하는 웹사이트와의 비교를 통해 규칙구성요소를 자동화하여 추출 할 수 있다. 이러한 온툴로지를 기반으로 규칙을 식별하기 위해서는 문자열 비교 기법을 사용하게 된다. 하지만 단순한 문자열 비교 기법만으로는 규칙을 식별하는 데에 자연어 처리에 대한 한계가 있다. 이를 극복하기 위해 다음의 두 번째 방법을 사용하고자 한다. 두 번째 방법은 정형화되지 않은 정보들을 확장하여 사용하는 것이다. 우선 찾고자 하는 단어들의 원형을 찾기 위한 스테밍 알고리즘 기법, WordNet을 이용하여 동의어 유의어등으로 확장을 하는 WordNet Expansion 기법, 의미 유사도를 측정하기 위한 방법인 Semantic Similarity Measure 등을 단계적으로 수행하여 자동화되고 정확한 규칙식별을 하고자 한다. 이러한 방법들의 조합으로 인하여 규칙구성요소 추출이 되지 않을 후보 단어들의 수를 줄여서 보다 더 정확하고, 지능적인 규칙구성요소 추출 방법론을 제시하고 구현하여 지식관리자의 규칙습득에 대한 부담을 줄여 주고자 한다.

• The KIPS Transactions:PartD
• /
• v.13D no.3 s.106
• /
• pp.309-316
• /
• 2006

We propose a novel technique to categorize XML documents and extract a concept efficiently using domain ontology. First, we create domain ontology that use text mining technique and statistical technique. We propose a DScore technique to classify XML documents by using the structural characteristic of XML document. We also present TScore technique to extract a concept by comparing the association term set of domain ontology and the terms in the XML document. To verify the efficiency of the proposed technique, we perform experiment for 295 papers in the computer science area. The results of experiment show that the proposed technique using the structural information in the XML documents is more efficient than the existing technique. Especially, the TScore technique effectively extract the concept of documents although frequency of term is few. Hence, the proposed concept-based retrieval techniques can be expected to contribute to the development of an efficient ontology-based knowledge management system.

• Spatial Information Research
• /
• v.18 no.5
• /
• pp.83-92
• /
• 2010

Recently, as the Geospatial Web is combined with the Semantic Web in order to keep pace with the recent trends of information technology emphasizing interoperability, intelligence and individualization, the Geo Semantic Web was proposed, which is an intelligent geographical information Web service technology that can provide users with suitable information by connecting and integrating various types of spatial information and extensive aspatial information on the Web efficiently. For the Geo Semantic Web service, we need to develop Geo Ontology processing technologies that enable computers to process knowledge and information scattered around in the Web environment automatically. However, standards for Geo Ontology processing technologies have nod been established yet, and standardization organizations and various groups and agencies are conducting relevant studies. This paper analyzed various base theories and technologies related to Geo Ontology and developed a Geo Semantic Web system. The Geo Semantic Web system comprises Query Processing Manager that analyzes and processes Geo Semantic queries and manages sessions, Ontology Manager that generates and queries Geo Ontology and extracts spatial/aspatial data, and Clients. Finally, this paper proved the utility of the Geo Semantic Web system by applying it to a hypothetical scenario where Geo Semantic queries are required.

• Korean Journal of Construction Engineering and Management
• /
• v.13 no.3
• /
• pp.99-108
• /
• 2012

The rough design information is only available from BIM(Building Information Model) based schematic design. So, it is difficult to obtain sufficient information for generating BOQ. Like 2D design, there are some problems that the results are depend on what the choice of cost estimator. However, the most research of BIM based cost estimation are focus on quantity takeoff, the consideration of work item for generating BOQ is insufficient. Therefore, this paper present automatic inference process of work items in a BOQ using ontology. The proposed process and ontology is validated through applying tiling construction. If the proposed process is utilized, it is expected the basis of developing generation method for consistent BOQ by resolving intervention of cost estimator's arbitrary decision.

• Journal of the Korean Society for information Management
• /
• v.20 no.3
• /
• pp.241-259
• /
• 2003

The Basic Semantic Register(BSR) is and official ISO register designed for interoperability among eBusiness and EDI systems. The entities registered in the current BSR are not defined in a machine-understandable way, which renders automatic extraction of structural and relationship information from the register impossible. The purpose of this study is to offer a framework for designing an ontology that can provide semantic interoperability among BSR-based systems by defining data structures and relationships with RDF and OWL, similar meaning by the 'equivalentClass' construct in OWL, the hierachical relationships among classes by the 'subClassOf' construct in RDF schema, definition of any entities in BSR by the 'label' construct in RDF schema, specification of usage guidelines by the 'comment' construct in RDF schema, assignment of classes to BSU's by the 'domain' construct in RDF schema, specification of data types of BSU's by the 'range' construct in RDF schema. Hierarchical relationships among properties in BSR can be expressed using the 'subPropertyOf' in RDF schema. Progress in semantic interoperability can be expected among BSR-based systems through applications of semantic web technology suggested in this study.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.6
• /
• pp.1220-1228
• /
• 2009

Ontologies, which are the core of the Semantic Web, are usually limited by specific domains or created by defining meanings and relationships that depend on the heuristic. However, the creation of an ontology is not only very difficult but also very time-consuming. In contrast with ontologies that are used in specific fields, an ontology for the Web entails an unlimited scope of knowledge and expression of information. Hence, it is hard to express information in the same way that is used to create ontologies in specific fields. Therefore, the automatic generation of an ontology takes very important role in the Semantic Web. In this paper, to make ontologies automatically, we suggest the methods to create and renew ontologies by expanding keywords related to the index-terms which are extracted from the search keywords which users input in the search engines by analyzing the morphemes.

• Journal of KIISE:Databases
• /
• v.35 no.4
• /
• pp.333-344
• /
• 2008

In recent years, the web services composition has received much attention. By web services composition, we mean providing a new service that does not exist on the repository. In this paper, we propose a new system called PSR for web services composition search using a relational database. We also propose algorithms for pre-computing web services composition using joins and indices. We store ontologies from web services in RDBMS, so that the PSR system returns web services composition in order of similarity with user query through the degree of the ontology matching. We demonstrated that our pre-computing web services composition approach in RDBMS yields lower execution time and good scalability when handling a large number of web services and user queries.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.2
• /
• pp.212-217
• /
• 2009

There are rule-based learning method and statistic based learning method and so on which represent learning method for hierarchy relation between domain term. In this paper, we propose to leveling and similarity measure using the extended AHP of fuzzy term in Information system. In the proposed method, we extract fuzzy term in document and categorize ontology structure about it and level priority of fuzzy term using the extended AHP for specificity of fuzzy term. the extended AHP integrates multiple decision-maker for weighted value and relative importance of fuzzy term. and compute semantic similarity of fuzzy term using min operation of fuzzy set, dice's coefficient and Min+dice's coefficient method. and determine final alternative fuzzy term. after that compare with three similarity measure. we can see the fact that the proposed method is more definite than classification performance of the conventional methods and will apply in Natural language processing field.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.7
• /
• pp.35-47
• /
• 2011

This paper proposes a set of rules to automatically generate OWL ontologies from relational databases. The purpose of the rules is to allow semantic access to existing RDB data without any database schema transformation and data migration process. In other words, the rules help a RDBMS play as a web ontology repository as well. However, the use of the mapping rules between RDB and OWL proposed by other studies for the objective causes troubles as follows. First, databases including the tables with a specific structure can't be translated into OWL. Second, the process for extracting an OWL individual unnecessarily lead to database join operations, or several SQL queries. On the other hand, our rules is designed to prevent these problems, can generate OWL classes and properties from database schemas and can generate OWL individuals from the database instances. In addition, an ontology generated by our rules is an OWL 2 DL ontology.