English-Korean Machine Translation Research Articles

MASS와 상대 위치 표현을 이용한 영어-한국어 신경망 기계 번역

신경망 기계 번역(Neural Machine Translation)은 주로 지도 학습(supervised learning)을 이용하는 Sequence-to-Sequence 모델에 대한 연구가 이루어지고 있다. 그러나, 지도 학습 방법은 데이터가 부족한 경우에는 낮은 성능을 보이기 때문에, 최근에는 BERT와 MASS 같은 대량의 단일 언어 데이터 기반 사전학습(pre-training) 모델을 이용하여 미세조정(fine-tuning)을 하는 전이 학습(transfer learning) 방법이 자연어 처리 분야에서 주로 연구되고 있다. 본 논문에서는 언어 생성(language generation) 작업을 위한 사전학습 방법을 사용하는 MASS 모델을 영어-한국어 기계 번역에 적용하였다. 실험 결과, MASS 모델을 이용한 영어-한국어 기계 번역 모델의 성능이 기존 모델들보다 좋은 성능을 보였고, 추가로 MASS 모델에 상대 위치 표현 방법을 적용하여 기계 번역 모델의 성능을 개선하였다.

전처리 및 후처리에 의한 언어 차이 해소 능력을 개선한 영한 기계번역 시스템

The linguistic differences between English and Korean cause difficulties in text processing, especially in machine translation. An English-Korean machine translation system must effectively solve these linguistic differences in order to be used practically. In this paper, we propose pre-processing as a method of solving the linguistic differences. We present pre-processing tasks for the differences and some accompanying post-processing procedures. We explore the differences requiring the pre-processing, devise pre-processing procedures, and determine the proper position where the procedures can be applied during the translation process. In an experiment, we show the usefulness of the proposed pre-processing and post-processing tasks for improving translation quality.

영어 구문 분석의 효율 개선을 위한 3단계 구문 분석

The performance of an English-Korean machine translation system depends heavily on its English parser. The parser in this paper is a part of the rule-based English-Korean MT system, which includes many syntactic rules and performs the chart-based parsing. The parser generates too many structures due to many syntactic rules, so much time and memory are required. The rule-based parser has difficulty in analyzing and translating the long sentences including the commas because they cause high parsing complexity. In this paper, we propose the 3-phase parsing method with sentence segmentation to efficiently translate the long sentences appearing in usual. Each phase of the syntactic analysis applies its own independent syntactic rules in order to reduce parsing complexity. For the purpose, we classify the syntactic rules into 3 classes and design the 3-phase parsing algorithm. Especially, the syntactic rules in the 3rd class are for the sentence structures composed with commas. We present the automatic rule acquisition method for 3rd class rules from the syntactic analysis of the corpus, with which we aim to continuously improve the coverage of the parsing. The experimental results shows that the proposed 3-phase parsing method is superior to the prior parsing method using only intra-sentence segmentation in terms of the parsing speed/memory efficiency with keeping the translation quality.

Noun Sense Disambiguation using Co-Occurrence Relation in Machine Translation

Word Sense Disambiguation, the process of identifying the meaning of a word in a sentence when the word has multiple meanings, is a critical problem of machine translation. It is generally very difficult to select the correct meaning of a word in a sentence, especially when the syntactical difference between the source and target language is big, e.g., English-Korean machine translation. To achieve a high level of accuracy of noun sense selection in machine translation, we introduced a statistical method based on co-occurrence relation of words in sentences and applied it to the English-Korean machine translator RyongNamSan. ACM Computing Classification System (1998): I.2.7.

영한 기계번역 시스템의 영한 변환사전 확장 도구

영한 기계번역 시스템을 개발하기 위해서는 언어에 대한 다양한 정보를 필요로 하며, 특히 영어 단어에 대한 의미 정보를 포함하는 영한 변환사전의 풍부한 정보량은 번역품질에 중요한 요소이다. 지속적으로 생성되는 새로운 단어들은 사전에 등록되어 있지 않아 번역문에 영어 단어가 그대로 출력되어 번역품질을 저하시킨다. 또한 복합명사는 어휘분석, 구문분석을 복잡하게 하고 사전에 의미가 등록되지 않은 경우가 많아 올바르게 번역하기 어렵다. 따라서 영한 기계번역의 번역품질 향상을 위해서는 사전에 등록되어 있지 않은 단어들과 자주 사용되는 복합명사들을 수집하고 의미 정보를 추가하여 영한 변환사전을 지속적으로 확장하는 것이 필요하다. 본 논문에서는 인터넷 신문기사로부터 말뭉치를 추출하고, 사전 미등록 단어와 자주 나타나는 복합명사를 찾은 후, 이들에 대해 의미를 부착하여 영한 변환사전에 추가하는 일련의 과정으로 구성되는 영한 변환사전의 확장 방안을 제안하고 이를 지원하는 도구를 개발하였다. 사전 정보의 확대는 많은 사람의 노력을 필요로 하는 일이지만, 영한 기계번역 시스템의 개선을 위해서는 필수적이다. 본 논문에서 개발한 도구는 사람의 노력을 최소화 하면서, 영한 변환사전의 정보량 지속적인 확대를 위해 유용하게 활용되어 영한 기계번역 시스템의 번역품질 개선에 기여할 것으로 기대된다. Developing English-Korean machine translation system requires the construction of information about the languages, and the amount of information in English-Korean transfer dictionary is especially critical to the translation quality. Newly created words are out-of-vocabulary words and they appear as they are in the translated sentence, which decreases the translation quality. Also, compound nouns make lexical and syntactic analysis complex and it is difficult to accurately translate compound nouns due to the lack of information in the transfer dictionary. In order to improve the translation quality of English-Korean machine translation, we must continuously expand the information of the English-Korean transfer dictionary by collecting the out-of-vocabulary words and the compound nouns frequently used. This paper proposes a method for expanding of the transfer dictionary, which consists of constructing corpus from internet newspapers, extracting the words which are not in the existing dictionary and the frequently used compound nouns, attaching meaning to the extracted words, and integrating with the transfer dictionary. We also develop the tool supporting the expansion of the transfer dictionary. The expansion of the dictionary information is critical to improving the machine translation system but requires much human efforts. The developed tool can be useful for continuously expanding the transfer dictionary, and so it is expected to contribute to enhancing the translation quality.

효율적 영한기계번역을 위한 확률적 품사결정

Natural language processing has several ambiguity problems, and English-Korean machine translation especially includes those problems to be solved in each translation step. This paper focuses on resolving part-of-speech ambiguity of English words in order to improve the efficiency of English analysis, which is in part of efforts for developing practical English-Korean machine translation system. In order to improve the efficiency of the English analysis, the part-of-speech determination must be fast and accurate for being integrated with machine translation system. This paper proposes the probabilistic models for part-of-speech determination. We use Penn Treebank corpus in building the probabilistic models. In experiment, we present the performance of the part-of-speech determination models and the efficiency improvement of the machine translation system by the proposed part-of-speech determination method.

Intra-sentence segmentation based on support vector machines in English–Korean machine translation systems

This work is about intra-sentence segmentation performed before syntactic analysis of long sentences composed of at least 20 words in an English–Korean machine translation system. A long sentence has been known to spend enormous computational time and space when it is analyzed syntactically. It can also produce poor translation results. To resolve this problem, we partitioned a long sentence into a few segments to analyze each segment separately. To partition the sentence, firstly, we tried to find candidates for each segment position in the sentence. We then generated input vectors representing lexical contexts of the corresponding candidates and also used the support vector machines (SVM) algorithm to learn and recognize the appropriate segment positions. We used three kernel functions, the linear kernel, the polynomial kernel and the Gaussian kernel, to find optimal hyperplanes classifying proper positions and we compared results obtained from each kernel function. As a result of the experiments, we acquired 0.81, 0.83, and 0.79 f-measure values from the linear, polynomial and Gaussian kernel, respectively.

A Method for English-Korean Target Word Selection Using Multiple Knowledge Sources

Target word selection is one of the most important and difficult tasks in English-Korean Machine Translation. It effects on the overall translation accuracy of machine translation systems. In this paper, we present a new approach to Korean target word selection for an English noun with translation ambiguities using multiple knowledge such as verb frame patterns, sense vectors based on collocations, statistical Korean local context information and co-occurring POS information. Verb frame patterns constructed with dictionary and corpus play an important role in resolving the sparseness problem of collocation data. Sense vectors are a set of collocation data when an English word having target selection ambiguities is to be translated to specific Korean target word. Statistical Korean Local Context Information is an N-gram information generated using Korean corpus. The co-occurring POS information is a statistically significant POS clue which appears with ambiguous word. To evaluate our approach, we applied the method to Tellus-EK* system, English-Korean automatic translation system currently developed at ETRI [1],[2]. The experiment showed promising results for diverse sentences from web documents.

영한 기계번역에서 구문 분석 정확성 향상을 위한 구문 범주 예측

영한 기계번역에서 구문 분석 정확성 향상을 위한 구문 범주 예측

영한 기계 번역에서 미가공 텍스트 데이터를 이용한 대역어 선택 중의성 해소

본 논문에서는 미가공 말뭉치 데이터를 활용하여 영한 기계번역 시스템의 대역어 선택 시 발생하는 중의성을 해소하는 방법을 제안한다. 이를 위하여 은닉 의미 분석(Latent Semantic Analysis : LSA)과 확률적 은닉 의미 분석(Probabilistic LSA : PLSA)을 적용한다. 이 두 기법은 텍스트 문단과 같은 문맥 정보가 주어졌을 때, 이 문맥이 내포하고 있는 복잡한 의미 구조를 표현할 수 있다 본 논문에서는 이들을 사용하여 언어적인 의미 지식(Semantic Knowledge)을 구축하였으며 이 지식은 결국 영한 기계번역에서의 대역어 선택 시 발생하는 중의성을 해소하기 위하여 단어간 의미 유사도를 추정하는데 사용된다. 또한 대역어 선택을 위해서는 미리 사전에 저장된 문법 관계를 활용하여야 한다. 본 논문에서는 이러한 대역어 선택 시 발생하는 데이터 희소성 문제를 해소하기 위하여 k-최근점 학습 알고리즘을 사용한다. 그리고 위의 두 모델을 활용하여 k-최근점 학습에서 필요한 예제 간 거리를 추정하였다. 실험에서는, 두 기법에서의 은닉 의미 공간을 구성하기 위하여 TREC 데이터(AP news)론 활용하였고, 대역어 선택의 정확도를 평가하기 위하여 Wall Street Journal 말뭉치를 사용하였다. 그리고 은닉 의미 분석을 통하여 대역어 선택의 정확성이 디폴트 의미 선택과 비교하여 약 10% 향상되었으며 PLSA가 LSA보다 근소하게 더 좋은 성능을 보였다. 또한 은닉 공간에서의 축소된 벡터의 차원수와 k-최근점 학습에서의 k값이 대역어 선택의 정확도에 미치는 영향을 대역어 선택 정확도와의 상관관계를 계산함으로써 검증하였다.젝트의 성격에 맞도록 필요한 조정만을 통하여 품질보증 프로세스를 확립할 수 있다. 개발 된 패키지의 효율적인 활용이 내조직의 소프트웨어 품질보증 구축에 투입되는 공수 및 어려움을 줄일 것으로 기대된다.도가 증가할 때 구기자 열수 추출 농축액은 <TEX>$1.6182{\sim}2.0543$</TEX>, 혼합구기자 열수 추출 농축액은 <TEX>$1.7057{\sim}2.1462{\times}10^7\;J/kg{\cdot}mol$</TEX>로 증가하였다. 이와 같이 구기자 열수 추출 농축액과 혼합구기자 열수 추출 농축액의 리올리지적 특성에 큰 차이를 나타내지는 않았다. security simultaneously.% 첨가시 pH 5.0, 7.0 및 8.0에서 각각 대조구의 57, 413 및 315% 증진되었다. 거품의 열안정성은 15분 whipping시, pH 4.0(대조구, 30.2%) 및 5.0(대조구, 23.7%)에서 각각 <TEX>$0{\sim}38.0$</TEX> 및 <TEX>$0{\sim}57.0%$</TEX>이었고 pH 7.0(대조구, 39.6%) 및 8.0(대조구, 43.6%)에서 각각 <TEX>$0{\sim}59.4$</TEX> 및 <TEX>$36.6{\sim}58.4%$</TEX>이었으며 sodium alginate 첨가시가 가장 양호하였다. 전체적으로 보아 거품안정성이 높은 것은 열안정성도 높은 경향이며, 표면장력이 낮으면 거품형성능이 높아지고, 비점도가 높으면 거품안정성 및 열안정성이 높아지는 경향이 있었다.protocol.eractions between application agents that are developed using different In this paper, we propose a new method utilizing only raw corpus without additional human effort for disambiguation of target word selection in English-Korean machine translation. We use two data-driven techniques; one is the Latent Semantic Analysis(LSA) and the other the Probabilistic Latent Semantic Analysis(PLSA). These two techniques can represent complex semantic structures in given contexts like text passages. We construct linguistic semantic knowledge by using the two techniques and use the knowledge for target word selection in English-Korean machine translation. For target word selection, we utilize a grammatical relationship stored in a dictionary. We use k- nearest neighbor learning algorithm for the resolution of data sparseness Problem in target word selection and estimate the distance between instances based on these models. In experiments, we use TREC data of AP news for construction of latent semantic space and Wail Street Journal corpus for evaluation of target word selection. Through the Latent Semantic Analysis methods, the accuracy of target word selection has improved over 10% and PLSA has showed better accuracy than LSA method. finally we have showed the relatedness between the accuracy and two important factors ; one is dimensionality of latent space and k value of k-NT learning by using correlation calculation.

통계 정보를 이용한 전치사 최적 번역어 결정 모델

This paper proposes a statistical model for the translation of prepositions in English-Korean machine translation. In the proposed model, statistical information acquired from unlabeled Korean corpora is used to choose the best translation from several possible translations. Such information includes functional word-verb co-occurrence information, functional word-verb distance information, and noun-postposition co-occurrence information. The model was evaluated with 443 sentences, each of which has a prepositional phrase, and we attained 71.3% accuracy. (Sungshin Women`s University)

English Syntactic Disambiguation Using Parser's Ambiguity Type Information

This paper describes a rule-based approach for syntactic disambiguation used by the English sentence parser in E-TRAN 2001, an English-Korean machine translation system. We propose Parser's Ambiguity Type Information (PATI) to automatically identify the types of ambiguities observed in competing candidate trees produced by the parser and synthesize the types into a formal representation. PATI provides an efficient way of encoding knowledge into grammar rules and calculating rule preference scores from a relatively small training corpus. In the overall scoring scheme for sorting the candidate trees, the rule preference scores are combined with other preference functions that are based on statistical information. We compare the enhanced grammar with the initial one in terms of the amount of ambiguity. The experimental results show that the rule preference scores could significantly increase the accuracy of ambiguity resolution.

Collocation Dictionary Optimization Using WordNetand k -Nearest Neighbor Learning

In machine translation, collocation dictionaries are important for selecting accurate target words. However, if the dictionary size is too large it can decrease the efficiency of translation. This paper presents a method to develop a compact collocation dictionary for transitive verb–object pairs in English–Korean machine translation without losing translation accuracy. We use WordNet to calculate the semantic distance between words, and k-nearest neighbor learning to select the translations. The entries in the dictionary are minimized to balance the trade-off between translation accuracy and time. We have performed several experiments on a selected set of verbs extracted from a raw corpus of over 3 million words. The results show that in real-time translation environments the size of a collocation dictionary can be reduced up to 40% of its original size without significant decrease in its accuracy.