• 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
• /
• 2018.10a
• /
• pp.470-473
• /
• 2018

최근 제안된 순환 신경망 기반 Encoder-Decoder 모델은 기계번역에서 좋은 성능을 보인다. 하지만 이는 대량의 병렬 코퍼스를 전제로 하며 병렬 코퍼스가 소량일 경우 데이터 희소성 문제가 발생하며 번역의 품질은 다소 제한적이다. 본 논문에서는 기계번역의 이러한 문제를 해결하기 위하여 단일-언어(Monolingual) 데이터를 학습과정에 사용하였다. 즉, 역-번역(Back-translation)을 이용하여 단일-언어 데이터를 가상 병렬(Pseudo Parallel) 데이터로 변환하는 방식으로 기존 병렬 코퍼스를 확장하여 번역 모델을 학습시켰다. 역-번역 방법을 이용하여 영-한 번역 실험을 수행한 결과 +0.48 BLEU 점수의 성능 향상을 보였다.

• The KIPS Transactions:PartB
• /
• v.13B no.3 s.106
• /
• pp.345-352
• /
• 2006

The practical English-Korean machine translation system should be able to translate long sentences quickly and accurately. The intra-sentence segmentation method has been proposed and contributed to speeding up the syntactic analysis. This paper proposes the syntactic category prediction method using decision trees for getting accurate parsing results. In parsing with segmentation, the segment is separately parsed and combined to generate the sentence structure. The syntactic category prediction would facilitate to select more accurate analysis structures after the partial parsing. Thus, we could improve the parsing accuracy by the prediction. We construct features for predicting syntactic categories from the parsed corpus of Wall Street Journal and generate decision trees. In the experiments, we show the performance comparisons with the predictions by human-built rules, trigram probability and neural networks. Also, we present how much the category prediction would contribute to improving the translation quality.

• Journal of the Korea Convergence Society
• /
• v.12 no.5
• /
• pp.23-29
• /
• 2021

Recent deep learning-based natural language processing studies are conducting research to improve performance by training large amounts of data from various sources together. However, there is a possibility that the methodology of learning by combining data from various sources into one may prevent performance improvement. In the case of machine translation, data deviation occurs due to differences in translation(liberal, literal), style(colloquial, written, formal, etc.), domains, etc. Combining these corpora into one for learning can adversely affect performance. In this paper, we propose a new Corpus Weight Balance(CWB) method that considers the balance between parallel corpora in machine translation. As a result of the experiment, the model trained with balanced corpus showed better performance than the existing model. In addition, we propose an additional corpus construction process that enables coexistence with the human translation market, which can build high-quality parallel corpus even with a monolingual corpus.

• ETRI Journal
• /
• v.42 no.1
• /
• pp.90-100
• /
• 2020

Named entity recognition (NER) continues to be an important task in natural language processing because it is featured as a subtask and/or subproblem in information extraction and machine translation. In Urdu language processing, it is a very difficult task. This paper proposes various deep recurrent neural network (DRNN) learning models with word embedding. Experimental results demonstrate that they improve upon current state-of-the-art NER approaches for Urdu. The DRRN models evaluated include forward and bidirectional extensions of the long short-term memory and back propagation through time approaches. The proposed models consider both language-dependent features, such as part-of-speech tags, and language-independent features, such as the "context windows" of words. The effectiveness of the DRNN models with word embedding for NER in Urdu is demonstrated using three datasets. The results reveal that the proposed approach significantly outperforms previous conditional random field and artificial neural network approaches. The best f-measure values achieved on the three benchmark datasets using the proposed deep learning approaches are 81.1%, 79.94%, and 63.21%, respectively.

• Journal of the Korean Society of Industry Convergence
• /
• v.23 no.3
• /
• pp.515-521
• /
• 2020

Deep learning methods have been effectively used to provide great improvement in various research fields such as machine learning, image processing and computer vision. One of the most frequently used deep learning methods in image processing is the convolutional neural networks. Compared to the traditional artificial neural networks, convolutional neural networks do not use the predefined kernels, but instead they learn data specific kernels. This property makes them to be used as feature extractors as well. In this study, we compared the quality of CNN features for traditional texture feature extraction methods. Experimental results demonstrate the superiority of the CNN features. Additionally, the recognition process and result of a pioneering CNN on MNIST database are presented.

• The Journal of the Convergence on Culture Technology
• /
• v.4 no.2
• /
• pp.197-204
• /
• 2018

One of the hottest issues in our 21st century is AI. Just as the automation of manual labor has been achieved through the Industrial Revolution in the agricultural society, the intelligence information society has come through the SW Revolution in the information society. With the advent of Google 'Alpha Go', the computer has learned and predicted its own machine learning, and now the time has come for the computer to surpass the human, even to the world of Baduk, in other words, the computer. Machine learning ML (machine learning) is a field of artificial intelligence. Machine learning ML (machine learning) is a field of artificial intelligence, which means that AI technology is developed to allow the computer to learn by itself. The time has come when computers are beyond human beings. Many companies use machine learning, for example, to keep learning images on Facebook, and then telling them who they are. We also used a neural network to build an efficient energy usage model for Google's data center optimization. As another example, Microsoft's real-time interpretation model is a more sophisticated translation model as the language-related input data increases through translation learning. As machine learning has been increasingly used in many fields, we have to jump into the AI industry to move forward in our 21st century society.

• Journal of the Korea Convergence Society
• /
• v.12 no.12
• /
• pp.65-74
• /
• 2021

Recently, studies on Korean spelling correction have been actively conducted based on machine translation and automatic noise generation. These methods generate noise and use as train and data set. This has limitation in that it is difficult to accurately measure performance because it is unlikely that noise other than the noise used for learning is included in the test set In addition, there is no practical error type standard, so the type of error used in each study is different, making qualitative analysis difficult. This paper proposes new 'error type classification' for deep learning-based Korean spelling correction research, and error analysis perform on existing commercialized Korean spelling correctors (System A, B, C). As a result of analysis, it was found the three correction systems did not perform well in correcting other error types presented in this paper other than spacing, and hardly recognized errors in word order or tense.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.7
• /
• pp.1792-1802
• /
• 1996

The analysis of prepositional phrases in English-to Korean machine translation has problem on the PP-attachment resolution, semantic analysis, and acquisition of information. This paper presents an analysis system for prepositional phrases, which solves the problem. The analysis system consists of the PP-attachment resolution hybrid system, semantic analysis system, and semantic feature generator that automatically generates input information. It provides objectiveness in analyzing prepositional phrases with the automatic generation of semantic features. The semantic analysis system enables to generate natural Korean expressions through selection semantic roles of prepositional phrases. The PP-attachment resolution hybrid system has the merit of the rule-based and neural network-based method.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.221-224
• /
• 2019

딥러닝을 이용한 Neural Machine Translation(NMT)의 등장으로 기계번역 분야에서 기존의 규칙 기반,통계기반 방식을 압도하는 좋은 성능을 보이고 있다. 본 논문은 기계번역 모델도 중요하지만 무엇보다 중요한 것은 고품질의 학습데이터를 구성하는 일과 전처리라고 판단하여 이에 관련된 다양한 실험을 진행하였다. 인공신경망 기계번역 시스템의 학습데이터 즉 병렬 코퍼스를 구축할 때 양질의 데이터를 확보하는 것이 무엇보다 중요하다. 그러나 양질의 데이터를 구하는 일은 저작권 확보의 문제, 병렬 말뭉치 구축의 어려움, 노이즈 등을 이유로 쉽지 않은 상황이다. 본 논문은 고품질의 학습데이터를 구축하기 위하여 병렬 코퍼스 필터링 기법을 제시한다. 병렬 코퍼스 필터링이란 정제와 다르게 학습 데이터에 부합하지 않다고 판단되며 소스, 타겟 쌍을 함께 삭제 시켜 버린다. 또한 기계번역에서 무엇보다 중요한 단계는 바로 Subword Tokenization 단계이다. 본 논문은 다양한 실험을 통하여 한-영 기계번역에서 가장 높은 성능을 보이는 Subword Tokenization 방법론을 제시한다. 오픈 된 한-영 병렬 말뭉치로 실험을 진행한 결과 병렬 코퍼스 필터링을 진행한 데이터로 만든 모델이 더 좋은 BLEU 점수를 보였으며 본 논문에서 제안하는 형태소 분석 단위 분리를 진행 후 Unigram이 반영된 SentencePiece 모델로 Subword Tokenization를 진행 하였을 시 가장 좋은 성능을 보였다.