• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.67-70
• /
• 2017

상호(商號)란 상인이나 회사가 영업 활동을 위해 자기를 표시하는데 쓰는 명칭을 말한다. 일반적으로 국내 기업의 상호 표기법은 한글과 로마자를 혼용함으로 상호 검색 시스템에서 단어 불일치 문제를 발생시킨다. 본 연구에서는 이러한 단어 불일치 문제를 해결하기 위해 Sequence-to-sequence 모델을 이용하여 로마자 상호를 이에 대응하는 한글 상호로 변환하고 그 후보들을 생성하는 시스템을 제안한다. 실험 결과 본 연구에서 구축한 시스템은 57.82%의 단어 정확도, 90.73%의 자소 정확도를 보였다.

• Journal of KIISE
• /
• v.44 no.12
• /
• pp.1269-1274
• /
• 2017

This paper focuses on the Hangul-to-roman conversion of Korean names. The proposed method recognizes existing notation and provides results according to the frequency of use. There are two main reasons for the diversity in Hangul-to-roman name conversion. The first is the indiscreet use of varied notation made domestically and overseas. The second is the customary notation of current notation. For these reasons, it has become possible to express various Roman characters in Korean names. The system constructs and converts data from 4 million people into a statistical dictionary. In the first step, the person's name is judged through a process matching the last name. In the second step, the first name is compared and converted in the statistical dictionary. In the last step, the syllables in the name are compared and converted, and the results are ranked according to the frequency of use. This paper measured the performance compared to the existing service systems on the web. The results showed a somewhat higher performance than other systems.

• The Science & Technology
• /
• v.32 no.4 s.359
• /
• pp.12-12
• /
• 1999

로마자 표기법은 우리말을 한글이 아닌 로마자로 적는 법을 말한다. 현재 우리가 쓰고 있는 로마자 표기법은 발음 위주의 표기를 하고 부호가 잡다하여 컴퓨터로 쉽게 입력할 수 없는 단점을 갖고 있다. 그래서 필자는 간편하고 능률적인 국제 정보통신용 로마자 표기법을 제안한다.

• Journal of Korean Library and Information Science Society
• /
• v.36 no.4
• /
• pp.27-44
• /
• 2005

Over sixty Korean romanization systems have introduced in the course of history since 1800, but not most of them have experienced the test of proper duration. At present two major systems are introduced and focused in the bibliographic fields : Mccune-Reischauer system and Ministry of Culture & Tourism system(2000). Few libraries nay be driven to the new romanization schemes, however library users require a practical and easy-to-use Korean romanization system for searching the Korean items. From the view of bibliographic control for Korean information resources, the advantage and disadvantages for converting romanization systems have been shown as the result of the research.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.9
• /
• pp.2271-2280
• /
• 1998

In this paper, we propose a method to retneve english written korcan names efficientl, and design a multilingual name retrieval system, It is very difficult to retrieve english-written korean names in typical IR sytems. For example, "홍길동" is written in english as vanous forms such like "Hong, gildong", "Gildong Hong", "Hong kil dong", "Hong kil dong" and so on, We not only propose a rule-based querv expansion method to retrieve english-written korean names efficiently but also design a multiligual name retneval system which is consisted of query classifier, exception handler, query expander, query executor, exception list and rulebase, Finally we will try to show that english-written korean names could be efficiently retrieved with rule based name generator.

• Journal of Information Management
• /
• v.43 no.2
• /
• pp.199-222
• /
• 2012

Due to the increase of international business and activities, Koreans nowadays have higher needs to present their personal identity to the foreigners. In this process, the first requirement is to exchange personal names with foreigners. Therefore, the phonetic translation of Korean names into Roman alphabetic notation is frequently required, in order to deliver Korean personal names to the people who do not understand Korean alphabet. However, some confusions have been witnessed in the way of transforming Korean names into Roman (English) alphabet notation, due to the fact that there are many different ways to put Korean pronunciation into Roman (English) alphabet. This study examines different formats of Romanization of Korean personal names to find and suggest an optimal one. It first examines structures of and differences between Korean and Western personal names and usage patterns, reviews the issues surrounding Romanization of Korean personal names, and patternizes diverse Romanization formats currently used. Based on these examinations and consequent findings, I would like to suggest a format for the Romanization of Korean personal names which is considered to be the best.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.378-380
• /
• 2000

본 논문은 한국어 로마자의 표준안 확립을 위하 로마자 변환 테이블을 정리 및 종합하고 이를 실제 변환기로 개발하는데 필요한 과정에 대해 기술하였다. 테이블 구성을 위해 국어의 표준 발음의 전체 구성을 분석하고, 불필요한 규칙이 있거나 없을 경우의 규칙을 추가, 삭제하여, 상충이 있을 경우 우선 순위에 의해 규칙을 선택하여 규칙을 수정하여 적용하였다. 이렇게 생성된 표준 발음 테이블은 한글을 받아 적용한 수 표기별 자음과 모음의 테이블을 통해 로마자로 변환한다. 일관성 없는 규칙, 표준화의 설득력 및 홍보의 미숙등과 같은 과제로 활용의 예가 적었던 한국어 로마자 변환기는 공공기관의 지명 및 문화재 로마자의 표준화 등에 사용될 수 있으며, 인터넷을 통한 웹 로마자 사전 제작을 통하여 표준화의 문제에 가장 중요한 객관적인 지표로서 활용될 수 있다.[3]

• Journal of the Korea Society of Computer and Information
• /
• v.9 no.4 s.32
• /
• pp.71-76
• /
• 2004

In this paper, a web-based automatic transcription component is designed and implemented for mechanical conversion of revised Korean-to-Romanization notation rule. Specially. we proposes system architecture and algorithms that transcript Korean to Roman automatically after transliterate Korean to phonetic symbol applying phonological principles. The components operate under the web server's script mechanism. and the dictionary for exceptional usage is designed as an accessorial function supported either operating at web server internally or externally. The overall system architecture is presented by UML. specification and pseudo code. The proposed architecture can be implemented in encapsulated service by object oriented component and that can be easily adapted and modified on the internet environment and this system may have many advantages to improve efficiency, library reuse. extensibility at software development.

• Annual Conference on Human and Language Technology
• /
• 1993.10a
• /
• pp.437-447
• /
• 1993

국제 협약에 따라 한글에 관한 정보는 로마자로 표기해야 한다. 우리나라는 1959년 2월 로마자 표기법'을 제정, 수차례 개정을 통해 1983년 6월 문교부(Ministry Of Education)안을 발표했으나, 표지판, 역명, 교과서, 공공문서들에서 활용되었지만, 영자신문, 외국 학술지등은 M-R( Mccune-Reishauer ) 표기법을, 인명, 신문 및 방송매체 등은 혼합표기방식을 사용함으로써 인(人). 지명(地名), ID, 대표어등 정보 검색시 혼란을 야기시켰고, 개정안에 따를 표기 정정으로 수백억원에 가까운 예산을 낭비하였으며, 최근 ISO(International Standard Organization)에서는 남북한 단일화 및 기계적 변환을 요구하고 있으나, 반달표 표기곤란, 편리성 결여, 북한의 주장등의 사유로 제정등을 거론하고 있는 실정에 있다. 잘 쓰지 않는 ASCII 코드 중에서 반달점 폰트를 제작하고 단어, 문장, 문서를 STRING으로 받아 알고리즘화된 음운법칙을 적용하여 소리글자로 변환하고 MOE테이블에서 대응글자를 참조하여 기계적 변환이 가능하도록 하였으며, 세련된 디자인, 풀다운 팝업방식을 채택, 2HD 1장으로 국내최초로 개발하였다.

• KIISE Transactions on Computing Practices
• /
• v.24 no.2
• /
• pp.88-92
• /
• 2018

'UFO2MF' and 'Korean MetaFont Generator' have been developed to convert UFO codes to the corresponding Metafont codes by using outline editing methods. However, 'UFO2MF' cannot express smooth curves because it use only reference points without using control points. And 'Korean MetaFont Generator' can easily design the curve of characters, and supports Hangul characters, but this system does not support Roman characters. In this paper, we propose a UFO2xMF system, which can convert both Korean and Roman characters from outline text information into Metafont codes. UFO2xMF can apply attribute values which can change the shape of letters during the conversion of Metafont code. It is also a highly compatible system that can convert the characters of various languages not only Korean characters but also Roman and other characters into Metafont codes by applying letters with baseline and centerline of gravity.