• ETRI Journal
• /
• v.30 no.1
• /
• pp.155-157
• /
• 2008

This letter introduces a new method to automatically acquire paraphrases using bilingual corpora. It utilizes the bilingual dependency relations obtained by projecting a monolingual dependency parse onto the other language's sentence based on statistical alignment techniques. Since the proposed paraphrasing method can clearly disambiguate the sense of the original phrases using the bilingual context of dependency relations, it would be possible to obtain interchangeable paraphrases under a given context. Through experiments with parallel corpora of Korean and English language pairs, we demonstrate that our method effectively extracts paraphrases with high precision, achieving success rates of 94.3% and 84.6%, respectively, for Korean and English.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2008.10a
• /
• pp.90-93
• /
• 2008

In natural language, it is common that repetitive constituents in an expression are to be left out and it is necessary to figure out the constituents omitted at analyzing the meaning of the sentence. This paper is on recognition of boundaries of parallel noun phrases by figuring out constituents omitted. Recognition of parallel noun phrases can greatly reduce complexity at the phase of sentence parsing. Moreover, in natural language information retrieval, recognition of noun with modifiers can play an important role in making indexes. We propose an unsupervised probabilistic model that identifies parallel cores as well as boundaries of parallel noun phrases conjoined by a conjunctive particle. It is based on the idea of swapping constituents, utilizing symmetry (two or more identical constituents are repeated) and reversibility (the order of constituents is changeable) in parallel structure. Semantic features of the modifiers around parallel noun phrase, are also used the probabilistic swapping model. The model is language-independent and in this paper presented on parallel noun phrases in Korean language. Experiment shows that our probabilistic model outperforms symmetry-based model and supervised machine learning based approaches.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 2002.02a
• /
• pp.249-258
• /
• 2002

We propose an algorithm for the automatic acquisition of a bilingual lexicon in the legal domain. We make use of a parallel corpus of bilingual court judgments, aligned to the sentence level, and analyse the bilingual context profiles to extract corresponding legal terms in both languages. Our method is different from those in past studies as it does not require any prior knowledge source, and naturally extends to multi-word terms in either language. A pilot test was done with a sample of ten legal terms, each with ten or more occurrences in the data. Encouraging results of about 75% average accuracy were obtained. This figure does not only reflect the effectiveness of the method for bilingual lexicon acquisition, but also its potential for bilingual alignment at the word or expression level.

• International Journal of Control, Automation, and Systems
• /
• v.5 no.1
• /
• pp.99-103
• /
• 2007

Translation equivalence is very important for bilingual lexicography, machine translation system and cross-lingual information retrieval. Extraction of equivalences from bilingual sentence pairs belongs to data mining problem. In this paper, discriminative learning methods are employed to filter translation equivalences. Discriminative features including translation literality, phrase alignment probability, and phrase length ratio are used to evaluate equivalences. 1000 equivalences randomly selected are filtered and then evaluated. Experimental results indicate that its precision is 87.8% and recall is 89.8% for support vector machine.

• Annual Conference on Human and Language Technology
• /
• 1999.10e
• /
• pp.150-155
• /
• 1999

다국어를 이용한 통계적 자연어 처리의 연구가 진행됨에 따라 병렬 말뭉치의 중요성이 대두되고 있다. 그러나 여러 가지 제약점으로 인하여 현재 이용 가능한 한국어 병렬 말뭉치가 드문 상황이다. 월드 와이드 웹 상에는 다양한 언어로 번역된 문서들이 있으며 이를 병렬 말뭉치로 구축, 활용한다면 말뭉치의 희소성으로 인한 문제를 해결할 수 있다. 본 논문에서는 웹 상에서 번역문서 후보를 추출한 다음 HTML 문서 구조를 비교하여 번역문서인지를 판별하고 문장 단위 정렬을 이용하여 병렬 말뭉치로 구축하는 방법을 제시한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.11a
• /
• pp.161-164
• /
• 2006

인터넷이 발전하면서 웹에는 같은 내용을 다양한 언어로 표현한 문서들이 많이 존재한다. 이와 같은 웹 문서의 성질을 이용하여, 이 논문은 웹으로부터 수집된 병렬문서(parallel document)를 이용하여 한영 병렬말뭉치 구축 시스템을 설계하고 구현한다. 이 논문에서 구축과정을 요약하면 다음과 같다. 첫째, 웹 문서수집기를 이용해서 웹으로부터 한영 웹문서(html 문서)를 각각 수집한다. 둘째, 수집된 각 언어의 웹 문서에서 불필요한 내용(태그와 광고 문구 등)을 제거하여 문장을 추출하고, 추출된 문장을 단락단위로 정렬한다. 셋째, 단락단위로 정렬된 문서를 문장정렬(sentence alignment) 방법을 이용해서 문장을 정렬한다. 끝으로 정렬된 병렬문장을 단어 단위로 분리하여 병렬말뭉치를 구축한다. 이와 같은 방법으로 이 논문에서는 약 42만 5천 문장의 한영 병렬말뭉치를 구축하였다.

• Annual Conference on Human and Language Technology
• /
• 2010.10a
• /
• pp.130-135
• /
• 2010

문장 정렬은 두 개의 문서 간의 대응이 되는 문장을 찾는 작업이다. 이 방법은 통계적 기계 번역의 학습 문서인 병렬 말뭉치를 자동으로 구축하는데 필수적인 방법이다. 본 연구에서는 길이 정보에 추가적으로 유사도 정보를 반영하는 한영 문장 정렬 방법을 제안한다. 먼저 한국어로 된 문서를 기계번역 시스템에 적용하여 영어 문서로 변환한다. 그리고 번역된 영어로 된 문서 결과와 영어로 된 대상 문서 간의 정렬 작업을 수행한다. 정렬 완료된 결과와 원시 문서, 대상 문서로부터 최종적인 결과를 생성해낸다. 본 논문에서는 기계 번역을 이용하는 방법과 더불어 기존의 길이 기반 문장 정렬 프로그램에 문장 유사도 정보를 추가하여 단어 정렬의 성능 향상을 꾀하였다. 그 결과 "21세기 세종기획"의 최종 배포본 내에 포함된 한영 병렬 말뭉치에 대해 한영 문장 정렬 F-1 자질의 결과가 89.39%를 보였다. 이 수치는 기존의 길이 기반의 단어 정렬의 성능 평가 결과와 비교했을 때 약 8.5% 가량 성능이 향상되었다.

• Language and Information
• /
• v.9 no.2
• /
• pp.49-68
• /
• 2005

Application of chunking to English and some other European languages has shown that it is a viable parsing mechanism for natural languages. Although a small number of attempts have been made to apply chunking to the analysis of the Korean language, it still is not clear enough what criteria there are to identify appropriate units of chunking, and how efficient and valid the chunking algorithms would be when applied to some authentic Korean texts. The purpose of this research is to provide an alternative set of algorithms for chunking Korean, and to implement them, and to test them against some English-Korean parallel corpora, which is English and Korean bibles matched sentence by sentence. It is shown in the paper that aligning related texts and identifying matched phrases between the two languages can be achieved through appropriate chunking and matching algorithms defined on the morphologically-tagged parallel corpus. Chunking and matching processes are based on the content words rather than the function words, and the matching itself is done in terms of the transfer dictionary. The implementation is done in C and XML, and can be accessed through the Internet.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.127-146
• /
• 2022

Recently, as word embedding has shown excellent performance in various tasks of deep learning-based natural language processing, researches on the advancement and application of word, sentence, and document embedding are being actively conducted. Among them, cross-language transfer, which enables semantic exchange between different languages, is growing simultaneously with the development of embedding models. Academia's interests in vector alignment are growing with the expectation that it can be applied to various embedding-based analysis. In particular, vector alignment is expected to be applied to mapping between specialized domains and generalized domains. In other words, it is expected that it will be possible to map the vocabulary of specialized fields such as R&D, medicine, and law into the space of the pre-trained language model learned with huge volume of general-purpose documents, or provide a clue for mapping vocabulary between mutually different specialized fields. However, since linear-based vector alignment which has been mainly studied in academia basically assumes statistical linearity, it tends to simplify the vector space. This essentially assumes that different types of vector spaces are geometrically similar, which yields a limitation that it causes inevitable distortion in the alignment process. To overcome this limitation, we propose a deep learning-based vector alignment methodology that effectively learns the nonlinearity of data. The proposed methodology consists of sequential learning of a skip-connected autoencoder and a regression model to align the specialized word embedding expressed in each space to the general embedding space. Finally, through the inference of the two trained models, the specialized vocabulary can be aligned in the general space. To verify the performance of the proposed methodology, an experiment was performed on a total of 77,578 documents in the field of 'health care' among national R&D tasks performed from 2011 to 2020. As a result, it was confirmed that the proposed methodology showed superior performance in terms of cosine similarity compared to the existing linear vector alignment.

• Korean Journal of Cognitive Science
• /
• v.22 no.1
• /
• pp.57-78
• /
• 2011

Paraphrasing is the act of writing a text using other words without altering the meaning. Paraphrases can be used in many fields of natural language processing. In particular, paraphrases can be incorporated in machine translation in order to improve the coverage and the quality of translation. Recently, the approaches on paraphrase extraction utilize bilingual parallel corpora, which consist of aligned sentence pairs. In these approaches, paraphrases are identified, from the word alignment result, by pivot phrases which are the phrases in one language to which two or more phrases are connected in the other language. However, the word alignment is itself a very difficult task, so there can be many alignment errors. Moreover, the alignment errors can lead to the problem of selecting incorrect pivot phrases. In this study, we propose a method in paraphrase extraction that discriminates good pivot phrases from bad pivot phrases. Each pivot phrase is weighted according to its reliability, which is scored by considering the lexical and part-of-speech information. The experimental result shows that the proposed method achieves higher precision and recall of the paraphrase extraction than the baseline. Also, we show that the extracted paraphrases can increase the coverage of the Korean-English machine translation.