• 제목/요약/키워드: Pre-compiled Dictionary

검색결과 2건 처리시간 0.016초

한국어 형태소 분석을 위한 효율적 기분석 사전의 구성 방법 (Construction of an Efficient Pre-analyzed Dictionary for Korean Morphological Analysis)

  • 곽수정;김보겸;이재성
    • 정보처리학회논문지:소프트웨어 및 데이터공학
    • /
    • 제2권12호
    • /
    • pp.881-888
    • /
    • 2013
  • 기분석 사전은 형태소 분석기의 속도와 정확도를 향상시키고, 과분석을 줄이기 위해 사용된다. 하지만 기분석 사전에 저장된 어절 중에 저장된 형태소 분석 결과가 부족한 어절, 즉 불충분 분석 어절이 존재할 경우 오히려 형태소 분석기의 정확도를 떨어뜨리는 원인으로 작용할 수 있다. 본 논문에서는 세종 형태 분석 말뭉치(문어체, 2011)를 이용해 말뭉치의 크기와 어절 빈도의 변화에 따라 사전의 정답 제시율이 변화하는 양상을 측정하였다. 그리고 통계기반의 형태소 분석기인 SMA와 기분석 사전을 결합한 통합 시스템을 구성하여 기분석 사전의 충분 분석률이 99.82% 이상일 때 시스템 전체 성능이 향상되는 것을 확인하였다. 또한 160만 어절의 말뭉치를 이용할 때는 32회 이상 출현한 어절로, 630만 어절로 구성된 말뭉치를 이용할 때는 64회 이상 출현한 어절로 사전을 구성하는 것이 통합 시스템의 성능을 가장 높게 할 수 있었다.

실록(實錄) : 등록(謄錄)의 위계(位階) (The Sillok as National Supreme Archives : An archival interpretation)

  • 오항녕
    • 기록학연구
    • /
    • 제3호
    • /
    • pp.91-113
    • /
    • 2001
  • History always be re-interpreted as the time flows. 'The Sillok', Which was registered in Memory of World of UNESCO in 1997, is comprehensive documents of the Chosun Dynasty, which had been compiled after kings' death, The Sillok encompasses 473 years of the reign in their 848 volumes(1,893 chapters). It was a history itself and has been main source in studying Korean history. Due to the rise of studies on the Sillok, time has come to explore the nature of the Sillok and to criticize the text, which would be called 'The Sillok-Study'. In this context, this paper examined three concepts that categorize the nature of the Sillok as historical materials ;Is it book or record?; The Sillok in register system in pre-modern society; And the Sillok as the National Archives. Korean historians, including myself, haven't yet examined the question whether the Sillok is the Book or Record in terms of archival science. At first, I regarded it as history book, and with this presupposition, wrote several papers on the characteristics of the Sillok. However, I recognized that the Sillok are close to record rather that history book as I examined the definition of glossary of librarian study, OED (Oxford English Dictionary) and Encyclopedia of Britannica, etc. Definitely, the Sillok was neither compiled and published to be read and sold publicly, nor meant to the works of literature or scholarship. one may say that the court-historians wrote comments on the facts and therefore it was just scholarly work. However, because the court-historians produced their comments on their own businesses, the outcome of 'their scholarly works' were also records conceptually, as were daily court-journalists in Rome. Its publication also had a absolutely different meaning from that of modern society. It was a method to preserve the important national records and distributed each edition of them to plural repositories for its safety and security. How can we explain its book-like shape and the procedure of compilation after a kings' death. The answer is as follows ; In pre-modern society, it was a common record-keeping system in the world to register records materials in order to arrange the materials of different sizes and to store them conveniently. And the lack of scientific preservation or conservation skill also encouraged them to register original records. Actually, the court-historians who participated in the compiling process called themselves "registering officers". On the other hand, similar to social hierarchy, there was a hierarchical system of records, and the Sillok was placed at the top of this hierarchy. In conclusion, the Sillok was a kind of registered records in the middle ages and the supreme records in the records-world. In addition to this we can also conceptualize the Sillok as archives. Through the compiling process, the most important and valuable records were selected to be the parts of Sillok. This process corresponds to the modem records appraisal. In the next step, it was preserved in the Four Archives(史庫) which located at remote site as archives and only accessible by the descendents in the future, who might be the people of the next dynasty. And nobody could access or read the documents at that time except the authorized court-historians who were archivists of the Chosun Dynasty. From this perspective, I conclude that Sillok was the supreme confidential archives in the register system. I work for the Government Archives as a historian and archivist. Whenever I entered the exhibition hall of the Government Archives and Records Service(GARS) and saw the replica of the Archives of Taebeak Mountain built during Chosun period, I always asked to myself a question whether the Sillok can be a symbol of the archival tradition of Korea and the GARS. Now, I can say, 'Yes!' definitely.