RISS 학술연구정보서비스

검색
다국어 입력

http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.

변환된 중국어를 복사하여 사용하시면 됩니다.

예시)
  • 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
  • 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
닫기
    인기검색어 순위 펼치기

    RISS 인기검색어

      검색결과 좁혀 보기

      선택해제
      • 좁혀본 항목 보기순서

        • 원문유무
        • 원문제공처
          펼치기
        • 등재정보
        • 학술지명
          펼치기
        • 주제분류
        • 발행연도
          펼치기
        • 작성언어
        • 저자
          펼치기

      오늘 본 자료

      • 오늘 본 자료가 없습니다.
      더보기
      • 무료
      • 기관 내 무료
      • 유료
      • Access Methods for Distributed Directory Systems

        허남원,전길남 한국정보과학회 1986 한국정보과학회 학술발표논문집 Vol.13 No.1

        In this thesis, we designed access methods between the DSA's which are distributed among countries or organizations. We designed indirect access methods as well as direct access methods for the sake of user's convenience.

      • 복수어절 변환 단위의 인식 및 변환을 위한 사전적 접근방법

        허남원 啓明專門大學 産業開發硏究所 1999 啓明硏究論叢 Vol.17 No.2

        한국어와 일본어는 비슷한 문법 체계를 가지고 있어서 상당 부분 1:1 관계의 형태소 단위로 매칭이 이루어지고 있다. 따라서 한-일 기계번역 시스템은 비교적 간단한 직접 번역 방식만을 사용하여 서로 번역을 하더라도 높은 번역률을 얻을 수 있다. 그러나, 일반적인 관용구나 복합 형태소, 보조 용언 등 일부는 n:m의 관계로 매칭이 이루어지며, 또한, 구성 요소들이 서로 인접하여 나타나지 않는 경우도 발생한다. 이들은 복수어절 변환 단위로 처리되어야 하며, 따라서, 본 연구에서는 한-일 기계번역의 관점에서 복수어절 변환 단위를 정의하고 이들을 효과적으로 인식하고 적절한 목적 언어로 변환하기 위한 방법을 제안한다. 이를 위하여 한-일 기계번역 시스템에서 다루어져야 할 복수어절 변환 단위의 특성에 맞게 복수어절 변환 단위를 표현할 수 있는 인식 범위 지정자를 정의하고, 효과적인 인식을 위한 인식 규칙들을 제시함으로써 인식의 정확성을 꾀하고 인식 시간을 효과적으로 최적화시킬 수 있도록 한다. Many previous researches on Multi-Word Translation Unit(MWTU) have been done under the framework of semantic network, smart parser, informative lexicon and so on. However, none of them are practically applicable to the Korean-Japanese machine translation system, not only because they need syntactic and semantic information which is unavailable in the direct MT system but also because their lexicon is very hard to construct. To overcome the difficulties, this thesis proposes a lexicon-based approach, in which the recognition and the transfer of MWTU are performed through a transfer dictionary with Trie indexing for MWTUs. In an experimental evaluation, the proposed algorithm proves its validation and effectiveness.

      • 중소기업에서의 중역정보시스템 개발 프레임웍에 관한 연구

        허남원,박학희 啓明專門大學 産業開發硏究所 2002 啓明硏究論叢 Vol.20 No.1

        The goal of this study is to develop a strategy of the "D" leasing company and an experimental prototype as analyzing feasibility and necessity of implementing EIS(Executive Information System) in a Leasing Company through the investigation of the general EIS development process. We suggest an actual system as a result of literature survey and many interviews with people on the job. On the way of reaching our purpose of this study, we find the fact that it will be desirable to make more correct diagnoses on internal conditions and external environment and information systems of the given firm in establishing EIS appropriate the surrounds of the company. We also find that it will be desirable to obtain the more accurate information required by executives to establish more suitable EIS for a company.

      • 다단계 언어학적 접근을 이용한 연결어미 애매성 해소

        허남원 啓明專門大學 産業開發硏究所 2001 啓明硏究論叢 Vol.19 No.1

        본 논문에서는 일본어 접속어미의 자연스럽고 정확한 번역을 위하여 한일 양국어 사이의 비교 연구를 기초로 한 다중 단계 언어 접근법을 제시하였다. 특히, 제안된 방법은 일본어 연결어미 중에서도 가장 많이 출현하면서 한국어로 번역 할 경우 심각한 애매성을 보이는 {-て}의 번역에 적용하였다. 제안된 방법은 언어학적인 접근법으로 형태론적, 통사론적, 의미론적 접근을 기초로 하였다. 형태론적 접근법에서는 일본어 문장에서 나오는 특정한 단어나 형태를 이용하여 목적언어를 결정하였고, 통사론적 접근법에서는 문장구조와 품사를 이용하였으며 의미적 접근법에서는 선행 및 후행 술어의 의미를 비교하였다. 제안된 다중 단계 언어 접근법은 기존의 해석정확률을 25.5%이상 우회하는 79.0%의 정확률을 보였다. This paper propose a multi-level linguistic method for a more natural and accurate translation of Japanese connective endings into Korean, on the basis of their contrastive analysis between the two languages. Specifically, the proposed method put an emphasis on the Japanese connective ending {-て}, one of the most frequently appearing but most ambiguous connective endings, when it is translated into the Korean connective endings. The proposed approach is based on three linguistic considerations, morphological, syntactic, and semantic. The morphological approach uses special words or patterns that appear with the connective endings to help ascertain the correct target connective ending. The syntactical approach uses sentence construction and parts of speech while the semantic approach uses a referencing of predicate classifications in a comparison of the predicate preceding the connective ending with the predicate that follows. Our proposed multi-level linguistic methods showed a translation accuracy of over 79.0%, and improvement of the previous machine translation system by about 25.3%.

      • KCI등재
      • KCI등재

        SWAT ArcView GIS Extension Patch를 이용한 소유역 분할에 따른 수문 및 유사 거동에 미치는 영향 평가

        성구 ( Sung Gu Heo ),남원 ( Nam Won Kim ),박윤식 ( Youn Shik Park ),김종건 ( Jong Gun Kim ),김성준 ( Seong Joon Kim ),안재훈 ( Jae Hun Ahn ),김기성 ( Ki Sung Kim ),임경재 ( Kyoung Jae Lim ) 한국물환경학회 2008 한국물환경학회지 Vol.24 No.2

        Because of increased nonpoint source runoff potential at highland agricultural fields of Kangwon province, effective agricultural management practices are required to reduce the inflow of sediment and other nonpoint source pollutants into the water bodies. The watershed-scale model, Soil and Water Assessment Tool (SWAT), model has been used worldwide for developing effective watershed management. However, the SWAT model simulated sediment values are significantly affected by the number of subwatershed delineated. This result indicates that the SWAT estimated watershed characteristics from the watershed delineation process affects the soil erosion and sediment behaviors. However, most SWAT users do not spend time and efforts to analyze variations in sediment estimation due to watershed delineation with various threshold value although topography falsification affecting soil erosion process can be caused with watershed delineation processes. The SWAT model estimates the field slope length of Hydrologic Response Unit (HRU) based on average slope of subwatershed within the watershed. Thus the SWAT ArcView GIS Patch, developed by using the regression relationship between average watershed slope and field slope length, was utilized in this study to compare the simulated sediment from various watershed delineation scenarios. Four watershed delineation scenarios were made with various threshold values (700 ha, 300 ha, 100 ha, and 75 ha) and the SWAT estimated flow and sediment values were compared with and without applying the SWAT ArcView GIS Patch. With the SWAT ArcView GIS Patch applied, the simulated flow values are almost same irrespective of the number of subwatershed delineated while the simulated flow values changes to some extent without the SWAT ArcView GIS Patch applied. However when the SWAT ArcView GIS Patch applied, the simulated sediment values vary 9.7% to 29.8% with four watershed delineation scenarios, while the simulated sediment values vary 0.5% to 126.6% without SWAT ArcView GIS applied. As shown, the SWAT estimated flow and sediment values are not affected by the number of watershed delineation significant compared with the estimated flow and sediment value without applying the SWAT ArcView GIS Patch.

      • SCIESCOPUSKCI등재

        Cyclic Nona 및 Pentapeptide 유도체들의 합성과 단백질 분해 효소에 대한 저해활성 측정

        강신원,허남원,신홍대 ( Shin Won Kang,Nam Won Huh,Hong Dae Shin ) 생화학분자생물학회 1989 BMB Reports Vol.22 No.4

        Most of legume seeds contain serveral kinds of BBI-type inhibitors. They have similar amino acid residues around the reactive site. The P₁ site amino acid residue in reactive site generally corresponds to the specificity of the cognate enzyme. It has also been noted the importance of P₁ site amino acid residue in synthetic cyclic nonapeptide derivatives which were related to active site of BBI. To obtain further understanding of the role of amino acid residues near th reactive site, 3 cyclic nonapeptides which are substituted analogues of GBI-II were synthesized and the inhibition constants for some proteinases were calculated by Dixon method. Antielastic fragment in GBI-II (I_a) differ only P₂ Site with that of soybean isoinhibitor C-II, Ki of Ia for porcine pancreatic elastase was 48μM, Ki of antitryptic fragment in LBI (I_b) for trypsin was 3.7μM, and Tyr analogue (I_c) for chymotrypsin showed no inhibition but appeared low inhibitory activity for trypsin and elastase. To elucidate the relation between loop size and inhibitory activity, 3 cyclic pentapeptide analogues containing the each reactive site of Ala-Ser, Lys-Ser and Tyr-Ser were synthesized. These peptides possessed slight inhibitory activity toward proteinases, Ki of synthetic cyclic pentapeptides for elastase, trypsin, and chymotrypsin were 614μM, 278μM and 128μM, respectively.

      • SCOPUSKCI등재

        GBⅠ-Ⅱ 와 관련된 Cyclic Peptide 들의 합성과 단백질 분해 효소에 대한 저해활성 연구

        강신원,허남원,Gang, Sin Won,Heo, Nam Won 대한화학회 1990 대한화학회지 Vol.34 No.3

        Antielastic fragment in GBⅠ-Ⅱ differ on $P_1$ site with that of antitryptic fragment in LBI. To obtain further understanding of the role of amino acid residue near the reactive site, specificity of $P_1$ site and loop size, Tyr substituted cyclic nonapeptide and cyclic pentapeptide were synthesized and the inhibition constants for some proteinases were calculated by Dixon method. Cyclic nonapeptide showed no inhibition for chymotrypsin but appeared low inhibitory activity for trypsin and elastase and that of cyclic pentapeptide possessed inhibitory activity for chymotrypsin. GBⅠ-Ⅱ의 antielastic 고리과 LBI의 antitryptic 고리는 $P_1$ 위치의 아미노산 잔기만 차이가 있으며 나머지 모든 아미노산 잔기는 동일하다. Inhibitor의 $P_1$을 키모트립신에 효과적인 특이성이 있다고 알려진 Tyr 잔기로 치환시킨 cyclic nonapeptide와 저해작용에 필수적으로 생각되는 5개의 아미노산 잔기만으로 선정된 cyclic pentapeptide 유도체를 액상법으로 합성하였다. 이들 펩티드 유도체를 3종의 단백질 분해효소에 작용시켜 그 저해활성을 측정한 결과 cyclic nonapeptide는 키모트립신에는 저해작용이 없었고 의외로 에라스타제와 트립신에 저해활성이 있었다. 또한 cyclic pentapeptide는 키모트립신에만 저해활성이 있었다.

      연관 검색어 추천

      이 검색어로 많이 본 자료

      활용도 높은 자료

      해외이동버튼