• Journal of Korea Society of Industrial Information Systems
• /
• v.4 no.4
• /
• pp.40-46
• /
• 1999

This paper introduce a Koran-Japanese machine translation system which is a module in the spoken language interpreting system It is implemented based on the TDMT(Transfre Driven Machine Translation). We define a new unit of translation so called TOKEN. The TOKEN-based translation method resolves nonstructural feature in Korean sentences and increases the quaity of translating results. In our system, we get rid of useless effort for traditional parsing by performing semi-parsing. The semi-parser makes the dependency tree which has minimum information needed generating module. We constructed the generation dictionaries by using the corpus obtained from ETRI spoken language database. Our system was tested with 600 utterances which is collected from travel planning domain The success-ratio of our system is 87% on restricted testing environment and 71% on unrestricted testing environment.

• Annual Conference on Human and Language Technology
• /
• 1999.10e
• /
• pp.249-254
• /
• 1999

자연어 분석에서 발생하는 가장 큰 문제점은 분석의 각 단계에서 필요 이상의 모호성이 발생하는 것이다. 이러한 모호성은 각각의 분석 단계에서는 반드시 필요한 결과일 수 있지만 다음 단계의 관점에서는 불필요하게 과생성된 자료로 볼 수 있다. 특히 한국어 형태소 분석 단계는 주어진 문장에 대해 최소의 의미를 가지는 형태소로 분석하기 때문에 과생성된 결과를 많이 만들어 내는데, 이들 대부분이 보조용언이나 의존 명사를 포함하는 형태소열에서 발생한다. 품사 태깅된 코퍼스에서 높은 빈도를 나타내는 형태소들을 분석해 보면 주위의 형태소와 강한 결합 관계를 가지는 것을 발견할 수 있다. 이러한 형태소는 대부분 자립성이 없는 기능형태소로서, 개개의 형태소가 가지는 의미의 합으로 표현되기보다는 문장내에서 하나의 구문 단위로 표현될 수 있다. 본 논문에서는 이 형태소 열을 구문 형태소로 정의하고, 필요한 경우 일반 형태소 해석의 결과를 구문 형태소 단위로 결합하고 이를 바탕으로 구문 해석을 하는 방법을 제안한다. 구문 형태소 단위를 이용하여 구문해석을 수행함으로써, 형태소 해석 결과의 축소를 통해 불필요한 구문 해석 곁과를 배제할 수 있다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.9 no.3
• /
• pp.64-70
• /
• 2004

Korean and Japanese have same structure of sentences because they belong to same family of languages. So, The transfer driven machine translation is most efficient to translate each other. This paper introduce a method which creates a transfer dictionary for Token Based Transfer Driven Koran-Japanese Machine Translation(TB-TDMT). If the transfer dictionaries are created well, we get rid of useless effort for traditional parsing by performing shallow parsing. The semi-parser makes the dependency tree which has minimum information needed output generating module. We constructed the transfer dictionaries by using the corpus obtained from ETRI spoken language database. Our system was tested with 900 utterances which are collected from travel planning domain. The success-ratio of our system is $92\%$ on restricted testing environment and $81\%$ on unrestricted testing environment.

• Journal of Broadcast Engineering
• /
• v.25 no.3
• /
• pp.346-352
• /
• 2020

Versatile Video Coding (VVC), a next-generation video coding standard that is in the final stage of standardization, has adopted various techniques to achieve more than twice the compression performance of HEVC (High-Efficiency Video Coding). VVC adopted Bi-prediction with CU-level Weight (BCW), which generates the final prediction signal with the weighted combination of bi-predictions with various weights, to enhance the performance of the bi-predictive inter prediction. The syntax element of the BCW index is adaptively coded according to the value of NoBackwardPredFlag which indicates if there is no future picture in the display order among the reference pictures. Such syntax structure for signaling the BCW index could violate the flexibility of video codec and cause the dependency issue at the stage of bitstream parsing. To address these issues, this paper proposes an efficient BCW weight signaling method which enables all weights and parsing without any condition check. The performance of the proposed method was evaluated with various weight searching methods in the encoder. The experimental results show that the proposed method gives negligible BD-rate losses and minor gains for 3 weights searching and 5 weights searching, respectively, while resolving the issues.