• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.05a
• /
• pp.397-398
• /
• 2007

본 논문에서는 환경변화에 대해 강인하게 동작하는 음성인식 시스템을 위해 잡음적응 훈련과 변별학습 방식을 결합한 형태의 환경적응 방식을 제안한다. 다중환경 훈련과 잡음제거방식을 결합한 형태인 잡음적응 훈련 방식은 음성인식을 위한 MCE (Minimum Classification Error)의 목적과는 거리가 있고, 음성인식 시스템이 사용되는 모든 환경을 반영하는 것은 현실적으로 어렵다는 점에서 한계가 있다. 이에 잡음적응 훈련방식으로 훈련된 기본 음향모델을 목적환경에서 수집한 소량의 데이터를 이용한 변별학습을 통해 환경적응 모델로 변환함으로써 이러한 단점을 보완할 수 있는 잡음 적응 변별학습을 이용한 훈련방식을 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.05a
• /
• pp.707-709
• /
• 2007

매일 생성되는 웹 신문에서 독자가 접해보지 못한 단어는 독자의 이해를 돕기 위하여 괄호를 사용한다. 괄호를 사용하여 표기된 웹 신문의 한국어-영어 대역쌍은 특정 기사에는 출현빈도가 낮지만 전체적으로 여러 신문의 기사를 봤을 때, 최소한 한번 이상 출현하게 된다. 즉, 괄호 안의 동일한 영어 용어 두 개 이상의 문장을 최장일치법 알고리즘에 적용하면 한국어 단어 경계를 자동으로 인식할 수 있다. 본 논문에서는 이런 웹 신문의 괄호 표기 특성을 이용하여 한-영 대역어쌍을 추출하는 방법을 제안한다. 웹 신문 기사 43,648 건에서 최대 2,087개의 한-영 대역어를 추출하였다. 3 개의 서로 다른 테스트 그룹으로 실험한 결과 최대 84.2%의 정확도를 보였다.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 2007.11a
• /
• pp.515-521
• /
• 2007

The text categorization is an important field for the automatic text information processing. Moreover, the authorship identification of a text can be treated as a special text categorization. This paper adopts the conceptual primitives' expression based on the Hierarchical Network of Concepts (HNC) theory, which can describe the words meaning in hierarchical symbols, in order to avoid the sparse data shortcoming that is aroused by the natural language surface features in text categorization. The KNN algorithm is used as computing classification element. Then, the experiment has been done on the Chinese text authorship identification. The experiment result gives out that the processing mode that is put forward in this paper achieves high correct rate, so it is feasible for the text authorship identification.

• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 1999.10c
• /
• pp.245-251
• /
• 1999

In this study, by using the natural language processing of the field artificial intelligence, automated index was performed . And then, the Natural Language Processor for Constructing Database (NALPDB) has been developed. Furthermore, the Design knowldege Information Relational DataBase (DIREDB) has been also developed, which is designed to interlock the knowledge base. DIREDB processes both the documented design-data , like a concrete standard specification, and the design knowledge frrom an expert. DIREDB is also simulates the design space of structures accordance with the production rule, and thus it is determined that DIREDB can be used as a engine to retrieve new knowledge and to implement knowldege base that is necessary to the development of automatic design system.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1744-1747
• /
• 2002

In late years, the tendency to shift the design language of electronic circuits from HDL to C-based languages of C/C)1 and so on is strengthened. The current of adopting these software languages thrives by necessity to solve the problem peculiar to HDL that verification of design is difficult. When we use C-based languages, we can describe the design by higher abstraction degree, mount the design as both hardware and software finally and so that express the design part which is not made clear at early stage the same one language. Therefore, the flexibility of design very improves, the design work in environment the range of applying the whole systems become possible. This paper introduces example at having applied C-based languages in image processing LSI design and describes that the design technique of C-based languages is effective for the system design.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10b
• /
• pp.1067-1070
• /
• 2000

WAP(Wireless Application Protocol)을 기반으로 한 무선 인터넷 분야의 발전과 더불어 엄청나게 늘어난 무선 인터넷 이용자들의 컨텐츠에 대한 요구가 다양해지고 있다. 따라서 CP(Content Provider)들은 무선 장비에 맞게 설계된 WML(Wireless Markup Language)과 WMLScipt를 무선 장비를 사용하지 않고, 실제와 비슷한 상황에서 컨텐츠(Content)를 개발할 수 있는 환경이 필요하게 되었다. 본 논문에서는 무선 웹 컨텐츠를 좀 더 쉽게 개발할 수 있도록 도와주는 WML 마이크로 브라우저의 구현에 대해 기술한다. WML 마이크로 브라우저는 WML 파싱(parsing) 모듈과 브라우징(Browsing)모들로 구성되고. 클라이언트 측 WAP 스택으로 요청한 결과로 얻은 컨텐츠를 화면에 출력한다.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.3
• /
• pp.792-803
• /
• 1999

Most data parallel language compilers are source-to-source translators. Most Compilers of HPF which is recognized as a standard data parallel language convert a parallel program in PHF in a Fortran 77 program inserted message passing primitives. By the way, they currently generate significant amount of ineffective codes in the course of the conversion. Especially, FORALL construct is converted into several DO loops, so loop overhead of these codes is very increased. In this paper, we define and use relation distance vector to keep necessary informations. Then we evaluate and analyze execution time for the codes converted by our method and by PARADIGM method for various array sizes.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.11
• /
• pp.2903-2913
• /
• 1997

This paper describes the design and implementation of the ParaC language that supports parallel programming on the shared memory and distributed memory parallel machine. The ParaC language is designed for the effective use of system resources of scalable parallel systems. The goal is achieved by adding parallel and synchronization constructs for shared address spaces, and remote task constructs for distributed address spaces. This paper also shows the translation method, and we implement the translator and the run-time library for parallel execution of extended constructs.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.12
• /
• pp.3539-3547
• /
• 1999

Korean linguistic information for the generation modulo of KANT(Knowledge-based Accurate Natural language Translation) system was constructed. As KANT has a language-independent generation engine, the construction of Korean linguistic information means the development of the Korean generation module. Constructed information includes concept-based mapping rules, category-based mapping rules, syntactic lexicon, template rules, grammar rules based on the unification grammar, lexical rules and rewriting rules for Korean. With these information in sentences were successfully and completely generated from the interlingua functional structures among the 118 test set prepared by the developers of KANT system.

• Language and Information
• /
• v.11 no.1
• /
• pp.77-96
• /
• 2007

This paper investigates syntactic variations in English such as Dative Alternation, Particle Inversion, and Object Postposition (Heavy NP Shift) within the framework of Optimality Theory, and shows that the same set of morphological, informational, and processing constraints affect all these variations. In particular, it shows that the variants that used to be regarded as ungrammatical are in fact used fairly often in reality, especially when processing or informational conditions are met, and therefore, grammatical judgment may not be always categorical but sometimes gradient. It is argued that the notion of distance in constraint ranking in stochastic OT can effectively explain the gradience and variability of grammaticality in the variation phenomena.