• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.629-632
• /
• 2021

기계 번역(machine translation)은 자연 언어로 된 텍스트를 다른 언어로 자동 번역 하는 기술로, 최근에는 주로 신경망 기계 번역(Neural Machine Translation) 모델에 대한 연구가 진행되었다. 신경망 기계 번역은 일반적으로 자기회귀(autoregressive) 모델을 이용하며 기계 번역에서 좋은 성능을 보이지만, 병렬화할 수 없어 디코딩 속도가 느린 문제가 있다. 비자기회귀(non-autoregressive) 모델은 단어를 독립적으로 생성하며 병렬 계산이 가능해 자기회귀 모델에 비해 디코딩 속도가 상당히 빠른 장점이 있지만, 멀티모달리티(multimodality) 문제가 발생할 수 있다. 본 논문에서는 단어 정렬(word alignment)을 이용한 비자기회귀 신경망 기계 번역 모델을 제안하고, 제안한 모델을 한국어-영어 기계 번역에 적용하여 단어 정렬 정보가 어순이 다른 언어 간의 번역 성능 개선과 멀티모달리티 문제를 완화하는 데 도움이 됨을 보인다.

• Communications for Statistical Applications and Methods
• /
• v.27 no.3
• /
• pp.313-326
• /
• 2020

Deep Learning is the most important key to the development of Artificial Intelligence (AI). There are several distinguishable architectures of neural networks such as MLP, CNN, and RNN. Among them, we try to understand one of the main architectures called Recurrent Neural Network (RNN) that differs from other networks in handling sequential data, including time series and texts. As one of the main tasks recently in Natural Language Processing (NLP), we consider Neural Machine Translation (NMT) using RNNs. We also summarize fundamental structures of the recurrent networks, and some topics of representing natural words to reasonable numeric vectors. We organize topics to understand estimation procedures from representing input source sequences to predict target translated sequences. In addition, we apply multiple translation models with Gated Recurrent Unites (GRUs) in Keras on English-Korean sentences that contain about 26,000 pairwise sequences in total from two different corpora, colloquialism and news. We verified some crucial factors that influence the quality of training. We found that loss decreases with more recurrent dimensions and using bidirectional RNN in the encoder when dealing with short sequences. We also computed BLEU scores which are the main measures of the translation performance, and compared them with the score from Google Translate using the same test sentences. We sum up some difficulties when training a proper translation model as well as dealing with Korean language. The use of Keras in Python for overall tasks from processing raw texts to evaluating the translation model also allows us to include some useful functions and vocabulary libraries as well.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 1995.02a
• /
• pp.195-204
• /
• 1995

We propose a method to improve search time and space complexity in statistical machine translation architectures, by employing linguistic bracketing information on the source language sentence. It is one of the advantages of the probabilistic formulation that competing translations may be compared and ranked by a. principled measure, but at the same time, optimizing likelihoods over the translation spa.ce dictates heavy search costs. To make statistical architectures practical, heuristics to reduce search computation must be incorporated. An experiment applying our method to a prototype Chinese-English translation system demonstrates substantial improvement.

• Language and Information
• /
• v.3 no.1
• /
• pp.65-82
• /
• 1999

The interests in Machine Translation(MT) have gotten revitalized lately with the rapid expansion of internet users. MT technology has gone through several different stages of development, but the longest surviving methods usually maintains the following characteristics: the expand ability and flexibility based on proved linguistic formalism, the transfer method of translation, the continued efforts of systematic updates being made into the system. This paper introduces one such system, L&H Korean-English bidirectional MT system. This system uses Lexical-Functional Grammar as its linguistic framework. It also adopts the transfer method of MT and has been around on the market for over 10 years for other language pairs. Currently, the system covers over 10 different languages including Chinese, Japanese and Arabic, in addition to European languages. This paper will review the system in its core and discuss related tools and resources be ing used to enhance the quality of translation.

• Journal of Korea Multimedia Society
• /
• v.7 no.7
• /
• pp.998-1008
• /
• 2004

There are many syntactic similarities between Japanese and Korean languages. Using these similarities, we can make out the Japanese-Korean translation system without most of syntactic analysis and semantic analysis. To improve the translation rates greatly, we have been developing the Japanese-Korean translation system using these similarities from several years ago. However, the system remains some problems such as a translation of inflected words, processing of multi-translatable words and so on. In this paper, we suggest the new method of Japanese-Korean translation by using relations of two neighboring words. To solve the problems, we investigated the connection rules of auxiliary verbs priority. And we design the translation table which is consists of entry tables and connection forms tables. A case of only one translation word, we can translate a Korean to Japanese by direct matching method use of only entry table, otherwise we have to evaluate the connection value by connection forms tables and then we can select the best translation word.

• ETRI Journal
• /
• v.32 no.5
• /
• pp.740-749
• /
• 2010

In a machine translation system, word sense disambiguation has an essential role in the proper translation of words when the target word can be translated differently depending on the context. Previous research on sense identification has mostly focused on adjacent words as context information. Therefore, in the case of nominal compounds, sense tagging of unit nouns mainly depended on other nouns surrounding the target word. In this paper, we present a practical method for the sense tagging of Korean unit nouns in a nominal compound. To overcome the weakness of traditional methods regarding the data sparseness problem, the proposed method adopts complement-predicate relation knowledge that was constructed for machine translation systems. Our method is based on a sentential form recovery technique, which recognizes grammatical relationships between unit nouns. This technique makes use of the characteristics of Korean predicative nouns. To show that our method is effective on text in general domains, the experiments were performed on a test set randomly extracted from article titles in various newspaper sections.

• Journal of the Korea Society of Computer and Information
• /
• v.5 no.4
• /
• pp.48-54
• /
• 2000

This Paper classified korean Parts-of-speech for korean-english machine translation and investigated morphological characters of each parts-of-speech. Korean standard grammar classified parts-of-speech by semantic, functional and formal character. Many rules make a difficulties the understanding of grammar structure and parts-of-speech classification and it is necessary to preprocess at machine translation. This paper classified korean parts-of-speech by one rule. The parts-of-speech suggested in this paper have a same syntactic role and same parts-of-speech with english dictionary, and express the structure of korean sentence. And also it can make target language by pattern matching in korean-english translation.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 1995.02a
• /
• pp.133-138
• /
• 1995

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.3
• /
• pp.681-687
• /
• 1998

The paper presents a hybrid mcthod for disambiguation of the verb meaning in the machine translation. The presented verb translation algorithm is to perform the concept-based method and the statistics-based method simultaneously. It uses a collocation dictionary, WordNct and the statistical information extracted from corpus. In the transfer phase of the machine translation, it tries to find the target word of the source verb. If it fails, it refers to Word Net to try to find it by calculating word similarities between the logical constraints of the source sentence and those in the collocation dictionary. At the same time, it refers to the statistical information extracted from corpus to try to find it by calculating co-occurrence similarity knowledge. The experimental result shows that the algorithm performs more accurate verb translation than the other algorithms and improves accuracy of the verb translation by 24.8% compared to the collocation-based method.

• Proceedings of the IEEK Conference
• /
• 2001.06c
• /
• pp.203-206
• /
• 2001

In the EKMT system, the part of Korea generation makes Korea sentence by using information obtained in the part of transfer. In the case of Korea generation, the conventional EKMT system don't arrange hierarchical word order and performs word order in the only modifier word. This paper proposes Korean adverb odering rule in English-Korean Machine Translation system which generates Korean sentence.