• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.504-509
• /
• 2018

Neural machine translation (NMT) has recently achieved the state-of-the-art performance. However, it is reported failing in the word sense disambiguation (WSD) for several popular language pairs. In this paper, we explore the extent to which NMT systems are able to disambiguate the Korean homographs. Homographs, words with different meanings but the same written form, cause the word choice problems for NMT systems. Consistent with the popular language pairs, we discover that NMT systems fail to translate Korean homographs correctly. We provide a Korean word sense disambiguation tool-UTagger to use for improvement of NMT's translation quality. We conducted translation experiments using Korean-English and Korean-Vietnamese language pairs. The experimental results show that UTagger can significantly improve the translation quality of NMT in terms of the BLEU, TER, and DLRATIO evaluation metrics.

• Journal of the Korea Convergence Society
• /
• v.11 no.6
• /
• pp.281-288
• /
• 2020

This study delves into acceptance on neural network machine translation (NMT) such as Google Translate and Papago that uses technical acceptance model. In conclusion, it turned out that perceived usefulness impacts translators' attitude towards NMT. In other words, if translators determine that NMT is related to their work and the quality of the deliverables is guaranteed, they were more positive towards it. Unlike the existing normative approach that translators feel threatened by NMT, empirical results tell us translators perceive NMT as a business tool and such perception was largely influenced by advices of their colleagues and friends and expectations for use.

• Phonetics and Speech Sciences
• /
• v.4 no.3
• /
• pp.137-142
• /
• 2012

The aim of this study is to investigate the usefulness of the multiple-choice name matching test (MC-NMT) in adults with aphasia by comparing the Korean version of the Boston Naming Test (K-BNT) and subsets of the Korean version of the Western Aphasia Battery (K-WAB). Thirty-nine patients who suffer from aphasia participated in the study. All patients were examined by the K-BNT, MC-NMT and K-WAB. The MC-NMT consisted of the 30 original BNT object stimuli which were presented with four response choices (written words) with similar frequency, including one correct and three incorrect responses. Cards containing the drawings were presented to the patient one at time. An item was passed if the patient chose the correct response within 10 seconds. We subdivided two groups into a total group and a low K-BNT group (at and below 15 points). We evaluated the correlation between the K-BNT, MC-NMT score and production, naming, repetition, comprehension, reading and writing scores in subsets of the K-WAB. There was a highly positive correlation between the K-BNT score and naming score of the K-WAB in total patients. However, the MC-NMT was highly correlated with reading scores in the K-WAB. In low score K-BNT patients, the K-BNT strongly correlated with production, naming and repetition scores of the K-WAB. These findings mean that K-BNT reflects motor language function. However, the MC-NMT was strong correlated comprehension, reading and writing of the K-WAB. This finding reflects sensory language function. We suggest that the combination of K-BNT and newly developed MC-NMT will be useful to evaluate speech functions in aphasic patients.

• Annual Conference on Human and Language Technology
• /
• 2015.10a
• /
• pp.115-118
• /
• 2015

Neural Machine Translation (NMT) 모델은 단일 신경망 구조만을 사용하는 End-to-end 방식의 기계번역 모델로, 기존의 Statistical Machine Translation (SMT) 모델에 비해서 높은 성능을 보이고, Feature Engineering이 필요 없으며, 번역 모델 및 언어 모델의 역할을 단일 신경망에서 수행하여 디코더의 구조가 간단하다는 장점이 있다. 그러나 NMT 모델은 출력 언어 사전(Target Vocabulary)의 크기에 비례해서 학습 및 디코딩의 속도가 느려지기 때문에 출력 언어 사전의 크기에 제한을 갖는다는 단점이 있다. 본 논문에서는 NMT 모델의 출력 언어 사전의 크기 제한 문제를 해결하기 위해서, 입력 언어는 단어 단위로 읽고(Encoding) 출력 언어를 문자(Character) 단위로 생성(Decoding)하는 방법을 제안한다. 출력 언어를 문자 단위로 생성하게 되면 NMT 모델의 출력 언어 사전에 모든 문자를 포함할 수 있게 되어 출력 언어의 Out-of-vocabulary(OOV) 문제가 사라지고 출력 언어의 사전 크기가 줄어들어 학습 및 디코딩 속도가 빨라지게 된다. 실험 결과, 본 논문에서 제안한 방법이 영어-일본어 및 한국어-일본어 기계번역에서 기존의 단어 단위의 NMT 모델보다 우수한 성능을 보였다.

• Magazine of korean Tunnelling and Underground Space Association
• /
• v.22 no.1
• /
• pp.26-35
• /
• 2020

Tunnelling to bypass major landslide areas is considered as a good and long-term environmentally friendly solution to reduce an existing hazard. In Norway, hundreds of kilometres of tunnels have been constructed in areas prone to landslides and snow avalanches. Although tunnelling is considered as an expensive mitigation strategy for bypassing landslides, analysis indicate that in some cases the cost of building a tunnel can be repaid by savings in driving costs (fuel) alone over a period of 5-10 years due to reduced driving distances. The other benefits of constructing tunnels in landslide areas include savings in time and increased safety. The Norwegian Method of Tunnelling (NMT) is considered safe, efficient and cost effective compared to other tunnelling techniques. Some aspects of NMT, which are considered safe and cost efficient, are presented. The application of updated rock support techniques, including reinforced ribs of shotctrete (RRS), which is a key component of the Norwegian Method of Tunnelling (NMT), is highlighted.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.141-142
• /
• 2021

최근 발전한 인공신경망 기반 기계 번역은 번역 시 더 자연스러운 번역을 제공한다. 본 논문에서는 기계번역기법을 이용하여 반말 표현을 존댓말 표현으로 변환하는 기법을 제안한다. 특히, 이를 위해 DCInside의 게시판을 크롤링하고 AI-HUB 데이터와 합쳐 약 20,000개의 자체 데이터 셋을 구축하였으며, 한글 전처리를 위한 4가지 기법 및 OpenNMT 프레임웍의 LSTM 및 Transformer 모듈을 활용하여 실험을 진행하였다. 이를 통해, 반말 표현을 높임 표현으로 변환하는 최적조합을 확인하였으며, 검증시 BLUE점수로 최대 66.53를 획득하였다.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.353-356
• /
• 2020

관용구는 둘 이상의 단어가 결합하여 특정한 뜻을 생성한 어구로 기계번역 시 종종 오역이 발생한다. 이는 관용구가 지닌 함축적인 의미를 정확하게 번역할 수 없는 기계번역의 한계를 드러낸다. 따라서 신경망 기계 번역(Neural Machine Translation)에서 관용구를 효과적으로 학습하려면 관용구에 특화된 번역 쌍 데이터셋과 학습 방법이 필요하다. 본 논문에서는 한-영 관용구 기계번역에 특화된 데이터셋을 이용하여 신경망 기계번역 모델에 관용구를 효과적으로 학습시키기 위해 특정 토큰을 삽입하여 문장에 포함된 관용구의 위치를 나타내는 방법을 제안한다. 실험 결과, 제안한 방법을 이용하여 학습하였을 때 대부분의 신경망 기계 번역 모델에서 관용구 번역 품질의 향상이 있음을 보였다.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.529-534
• /
• 2020

인공신경망 기반 기계번역(Neural Machine Translation, NMT)이란 딥러닝(Deep learning)을 이용하여 출발 언어의 문장을 도착 언어 문장으로 번역해주는 시스템을 일컫는다. NMT는 종단간 학습(end-to-end learning)을 이용하여 기존 기계번역 방법론의 성능을 앞지르며 기계번역의 주요 방법론으로 자리잡게 됐다. 이러한 발전에도 불구하고 여전히 개체(entity), 또는 전문 용어(terminological expressions)의 번역은 미해결 과제로 남아있다. 개체나 전문 용어는 대부분 명사로 구성되는데 문장 내 명사는 주체, 객체 등의 역할을 하는 중요한 요소이므로 이들의 정확한 번역이 문장 전체의 번역 성능 향상으로 이어질 수 있다. 따라서 본 논문에서는 지식그래프(Knowledge Graph)를 이용하여 심볼릭 지식을 NMT와 결합한 뉴럴심볼릭 방법론을 제안한다. 또한 지식그래프를 활용하여 NMT의 성능을 높인 선행 연구 방법론을 한영 기계번역에 이용할 수 있도록 구조를 설계한다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.1C
• /
• pp.45-54
• /
• 2010

This paper studies the design of a multiuser multiple-input multiple-output (MIMO) system, where a base station (BS) transmits independent messages to multiple users. The remarkable "dirty paper coding (DPC)" result was first presented by Costa that the capacity does not change if the Gaussian interference is known at the transmitter noncausally. While several implementable DPC schemes have been proposed recently for single-user dirty-paper channels, DPC is still difficult to implement directly in practical multiuser MIMO channels. In this paper, we propose a network channel matrix triangulation (NMT) algorithm for utilizing interference known at the transmitter. The NMT algorithm decomposes a multiuser MIMO channel into a set of parallel, single-input single-output dirty-paper subchannels and then successively employs the DPC to each subchannel. This approach allows us to extend practical single-user DPC techniques to multiuser MIMO downlink cases. We present the sum rate analysis for the proposed scheme. Simulation results show that the proposed schemes approach the sum rate capacity of the multiuser MIMO downlink at moderate signal-to-noise ratio (SNR) values.

• 도로교통
• /
• s.84
• /
• pp.40-44
• /
• 2001