• ETRI Journal
• /
• v.37 no.3
• /
• pp.541-550
• /
• 2015

In this paper, we propose a classification-based approach for hybridizing statistical machine translation and rulebased machine translation. Both the training dataset used in the learning of our proposed classifier and our feature extraction method affect the hybridization quality. To create one such training dataset, a previous approach used auto-evaluation metrics to determine from a set of component machine translation (MT) systems which gave the more accurate translation (by a comparative method). Once this had been determined, the most accurate translation was then labelled in such a way so as to indicate the MT system from which it came. In this previous approach, when the metric evaluation scores were low, there existed a high level of uncertainty as to which of the component MT systems was actually producing the better translation. To relax such uncertainty or error in classification, we propose an alternative approach to such labeling; that is, a cut-off method. In our experiments, using the aforementioned cut-off method in our proposed classifier, we managed to achieve a translation accuracy of 81.5% - a 5.0% improvement over existing methods.

• Language and Information
• /
• v.2 no.2
• /
• pp.185-220
• /
• 1998

The purpose of this research is to propose a set of basic guidelines for the construction of English test suites, a set of basic guidelines for the construction of Korean test suites to objectively evaluate the performance of machine translation systems. For this end, we constructed 650 English test sentences, 650 Korean test sentences, and developed the statistical methods and tools for the comparative evaluation of the English-Korean machine translation systems. It also evaluates the existing commercial English-Korean machine translation systems. The importance of this research lies in that it will promote an awareness of the importance and need of testing machine translation systems within the Natural Language Community. This research will also make a big contribution to the development of evaluation methods and techniques for appropriate test suites for Korean information processing systems. The results of this research can be used by the natural language community to test the performance and development of their information processing systems or machine translation systems.

• Journal of the Korea Management Engineers Society
• /
• v.23 no.4
• /
• pp.179-195
• /
• 2018

So far, the evaluation of machine translation has used a numerical approach, but there is evidence that it is not sufficient to reflect the characteristics or behavior of machine translation users(Hutchins, 2007; Wu et al., 2016; Park et al., 2013). Therefore, this study focused on the purpose of use and purchasing behavior of machine translation users. At this time, the indirect comparison method introduced by Morgan and Hunt(1994) was used to measure cognitive gaps and analyze the purchasing behavior of users. According to the analysis of association rules using cognitive gaps, the smaller the cognitive gap, the more positive the purchase behavior. In addition, procedural knowledge derived from language knowledge is activated in situations involving responsibility, and in routine situations, procedural knowledge trained from pragmatic knowledge works.

• 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.

• Korean Journal of Cognitive Science
• /
• v.12 no.4
• /
• pp.11-27
• /
• 2001

Machine translation (MT) quality assessment is an outstanding problem. In the present situation in which the quality of machine-translated products are far from the user\\`s satisfaction objective evaluation of MT system is a prerequisite to building mutual trust between the users and the vendors stimulating constructive competition among the developers and finally leading to improve the quality of MT systems. Especially there emerges a need for an intensive study on how to evaluate the quality of MT systems from both linguistic and data processing aspects and to secure a steady improvement of the translation quality. With due regard to such points we in this paper present a linguistic evaluation of English-to-Korean machine translation based on a test suite composed of 3.373 sentences that were classified into their linguistic phenomena and complexity levels and report the experimental results made from several commercial MT systems.

• Journal of the Korea Convergence Society
• /
• v.11 no.5
• /
• pp.1-8
• /
• 2020

Machine translation refers to a system where a computer translates a source sentence into a target sentence. There are various subfields of machine translation. APE (Automatic Post Editing) is a subfield of machine translation that produces better translations by editing the output of machine translation systems. In other words, it means the process of correcting errors included in the translations generated by the machine translation system to make proofreading. Rather than changing the machine translation model, this is a research field to improve the translation quality by correcting the result sentence of the machine translation system. Since 2015, APE has been selected for the WMT Shaed Task. and the performance evaluation uses TER (Translation Error Rate). Due to this, various studies on the APE model have been published recently, and this paper deals with the latest research trends in the field of APE.

• Language and Information
• /
• v.2 no.1
• /
• pp.79-105
• /
• 1998

This paper aims to describe a German-Korean machine translation system for nominal phrases. Besides, we have two subgoals. First, we are going to revea linguistic differences between two languages and propose a language-informational method fo overcome the differences. The method is based on an integrated model of translation knowledge, efficient information structure, and concordance selection. Then, we will show the statistical results about translation experiment and its evaluation as an evidence for the adequacy of our linguistic method and translation system itself.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 2007.11a
• /
• pp.105-114
• /
• 2007

This paper addresses a method for customizing an English-to-Korean machine translation system from general domain to patent domain. The customizing method consists of following steps: 1) linguistically studying about characteristics of patent documents, 2) extracting unknown words from large patent documents and constructing large bilingual terminology, 3) extracting and constructing the patent-specific translation patterns 4) customizing the translation engine modules of the existing general MT system according to linguistic study about characteristics of patent documents, and 5) evaluating the accuracy of translation modules and the translation quality. This research was performed under the auspices of the MIC (Ministry of Information and Communication) of Korean government during 2005-2006. The translation accuracy of the customized English-Korean patent translation system is 82.43% on the average in 5 patent fields (machinery, electronics, chemistry, medicine and computer) according to the evaluation of 7 professional human translators. In 2006, the patent MT system started an on-line patent MT service in IPAC (International Patent Assistance Center) under MOCIE (Ministry of Commerce, Industry and Energy) in Korea. In 2007, KIPO (Korean Intellectual Property Office) tries to launch an English-Korean patent MT service.

• Journal of Digital Contents Society
• /
• v.19 no.6
• /
• pp.1161-1167
• /
• 2018

The development of machine learning has enabled machines to perform delicate tasks that only humans could do, and thus many companies have introduced machine learning based translators. Existing translators have good performances but they have problems in number translation. The translators often mistranslate numbers when the input sentence includes a large number. Furthermore, the output sentence structure completely changes even if only one number in the input sentence changes. In this paper, first, we optimized a neural machine translation model architecture that uses bidirectional RNN, LSTM, and the attention mechanism through data cleansing and changing the dictionary size. Then, we implemented a number-processing algorithm specialized in number translation and applied it to the neural machine translation model to solve the problems above. The paper includes the data cleansing method, an optimal dictionary size and the number-processing algorithm, as well as experiment results for translation performance based on the BLEU score.

• Language and Information
• /
• v.2 no.1
• /
• pp.1-41
• /
• 1998

Evaluation of multilingual translation fundamentally involves measurement of meaning equivalences between the formally mapped discourses/texts of SL(source language) and TL(target language) both represented by a metalanguage called IL(interlingua). Unlike a usaal uni-directional MT(machine translation) model(e.g.:SL $\rightarrow$ analysis $\rightarrow$ transfer $\rightarrow$ generation $\rightarrow$ TL), a bi-directional(by 'negotiation') model(i.e.: SL $\rightarrow$ IL/S $\leftrightarrow$ IL $\leftrightarrow$ IL/T \leftarrow TL) is proposed here for the purpose of evaluating multilingual, not merely bilingual, translation. The IL, as conceived of in this study, is an English-based predicate logic represented in the framework of MRS(minimal recursion semantics), an MT-oriented off-shoot of HPSG(Head-driven Phrase Structure Grammar). In addition, a list of semantic and pragmatic checkpoints are set up, some being optional depending on the kind and use of the translation, so sa to have the evaluation of translation fine-grained by computing matching or mismatching of such checkpoints.