• Journal of the Korea Convergence Society
• /
• v.12 no.7
• /
• pp.37-44
• /
• 2021

Quality estimation (QE) can evaluate the quality of machine translation output even for those who do not know the target language, and its high utilization highlights the need for QE. QE shared task is held every year at Conference on Machine Translation (WMT), and recently, researches applying Pretrained Language Model (PLM) are mainly being conducted. In this paper, we conduct a survey on the QE task and research trends, and we summarize the features of PLM. In addition, we used a multilingual BART model that has not yet been utilized and performed comparative analysis with the existing studies such as XLM, multilingual BERT, and XLM-RoBERTa. As a result of the experiment, we confirmed which PLM was most effective when applied to QE, and saw the possibility of applying the multilingual BART model to the QE task.

• Journal of KIISE:Software and Applications
• /
• v.37 no.4
• /
• pp.297-304
• /
• 2010

Clustering method which based on sentence type or document genre is a technique used to improve translation quality of SMT(statistical machine translation) by domain-specific translation. But there is no previous research using sentence type and document genre information simultaneously. In this paper, we suggest an integrated clustering method that classifying sentence type by syntactic structure similarity and document genre by word similarity information. We interpolated domain-specific models from clusters with general models to improve translation quality of SMT system. Kernel function and cosine measures are applied to calculate structural similarity and word similarity. With these similarities, we used machine learning algorithms similar to K-means to clustering. In Japanese-English patent translation corpus, we got 2.5% point relative improvements of translation quality at optimal case.

• Journal of Korean Medical classics
• /
• v.23 no.5
• /
• pp.67-82
• /
• 2010

Globalization describes a process by which regional cultures have become integrated through a global network of communication. In order to communicate among different cultural groups, standardization of terminology is one of the most important steps among its processes. In the field of oriental medicine, there have been continuous efforts to communicate through various methods. Translation of oriental medical classics is one of the significant approaches in terms of transmitting medical theories and clinical experiences of thousands of years to the people of different cultural backgrounds. However, previous translation studies have had difficulties in delivering its underlying principles and assumptions due to lack of standardization of terminology. "WHO International Standard Terminologies on Traditional Medicine in the Western Pacific Region(WHO-IST)" is the outcome of developing standard terminologies on oriental medicine based on mutual agreement of researchers of Korea, China and Japan. As a movement to find more efficient methodology for communication between heterogeneous communities, this study aims to translate parts of "Classic of difficult issues(難經)" into English adopting "WHO-IST" hoping to set a model of translation study.

• Journal of Korea Multimedia Society
• /
• v.25 no.10
• /
• pp.1535-1545
• /
• 2022

Sign language has completely different meaning depending on the direction of the hand or the change of facial expression even with the same gesture. In this respect, it is crucial to capture the spatial-temporal structure information of each movement. However, sign language translation studies based on Sign2Gloss2Text only convey comprehensive spatial-temporal information about the entire sign language movement. Consequently, detailed information (facial expression, gestures, and etc.) of each movement that is important for sign language translation is not emphasized. Accordingly, in this paper, we propose Spatial-temporal Keypoints Centered Sign2Gloss2Text Translation, named STKC-Sign2 Gloss2Text, to supplement the sequential and semantic information of keypoints which are the core of recognizing and translating sign language. STKC-Sign2Gloss2Text consists of two steps, Spatial Keypoints Embedding, which extracts 121 major keypoints from each image, and Temporal Keypoints Embedding, which emphasizes sequential information using Bi-GRU for extracted keypoints of sign language. The proposed model outperformed all Bilingual Evaluation Understudy(BLEU) scores in Development(DEV) and Testing(TEST) than Sign2Gloss2Text as the baseline, and in particular, it proved the effectiveness of the proposed methodology by achieving 23.19, an improvement of 1.87 based on TEST BLEU-4.

• Korean Journal of Computational Design and Engineering
• /
• v.14 no.2
• /
• pp.96-104
• /
• 2009

It gradually becomes important to study on how to efficiently integrate and manage plant lifecycle data such as 2D schematic and 3D solid data, logical configuration data, and equipment specifications data. From this point of view, converting plant design data from various systems into neutral data independent from any commercial systems is one of important technologies for the operation and management of plants which usually have a very long period of life. In order to achieve this goal, a neutral model for efficient integration and management of plant data was defined. After schema mapping between one of specification-driven plant design systems and the neutral model was performed, a plant data translator is also implemented according to the mapping result. Finally, by experiments with nuclear power plant design, the feasibility of the translator was demonstrated.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2009.09a
• /
• pp.1011-1016
• /
• 2009

In this study, to performed laboratory model tests in order to verify on load transfer condition of drilled shaft under lateral loading. To conducted model test on polystyle drilled shaft under multi layer ground conditions. In model test, to measured the strain of drilled shaft and displacement under later loading. In order to verify on model test results, to conduct the numerical analysis.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.2 no.6
• /
• pp.94-101
• /
• 1994

About vibration model of Six-degrees-of-freedom(DOF), in mass load, examined results for knowing dynamic interference and response variation is as follows; In case of putting mass load upon the object, experimented results on two-degrees-of-freedom of the translation-1 direction and the rotation-1 direction at open-loop-control system, about 0.19 arcsed in input of the translation-$0.1{\mu}m$ and $0.022{\mu}m$ on input of the rotation-0.5 arcsec, the justicse of motion equation is acknowledged as confirming the appearance of the interference-$0.022{\mu}m$. In establishing calculation of transformation matrix by using analogue circuit, as simulating results that used incomplete differentiation, interference is $1.7{\times}10^{-3}$ arcsec on input of the translation-$0.1{\mu}m$ and $1.4{\times}10^{4}{\mu}m$ on input of the rotation-0.5 arcsec in open-loop-control system. Also it is $4.2{\times}10^{-4}$ arcsec on input of the translation-$0.1{\mu}m$ and $5.6{\times}10^{-5}{\mu}m$ on input of the rotation-0.5 arcesc in closed-loop-control system. As closed-loop-control system is better than open-loop-control system, equivalent accordance is confirmed on original response. Finally, fundamental validity of this theory is acknowledged.

• The Journal of Society for e-Business Studies
• /
• v.11 no.3
• /
• pp.73-88
• /
• 2006

As a computerized application for controlling enterprise processes, BPM(Business Process Management) has been received such concerns that many solution vendors developed their own process definition and storage methods. The fact causes the needs of process modeling standards for process model exchange and at the same time, requires the appropriate methodology for adopting the global standards. In this paper, we propose a transformation methodology of BPMN-based file into XMI(XML Metadata Interchange)-based neutral file format. We devised translation templates for 21 workflow patterns, and compared the results with BPEL4WS(Business Process Execution Language for Web Services) translation. As a result, our XMI transformation model enables more complete translation of process model in comparison with existing model, and thus can be practically utilized to the BPM vendors adopting BPMN standards.

• Restorative Dentistry and Endodontics
• /
• v.20 no.2
• /
• pp.655-669
• /
• 1995

Restorative procedures can lead to weakening tooth due to reduction and alteration of tooth structure. It is essential to prevent fractures to conserve tooth. Among the several parameters in cavity designs, cavity isthmus is very important. In this study, amalgam 0 cavity was prepared on maxillary first premolar. Two dimensional finite element models were made by serial photographic method and isthmus(1/4, 1/3, 1/2, 2/3 of intercuspal distance) were varied. Three or four-nodal mesh were used for the two dimensional finite element models. The periodontal ligament and alveolar bone surrounding the tooth were excluded in these models. 1S model was sound tooth with no amalgam cavity. B model was assumed perfect bonding between the restoration and cavity wall. Both compressive and tensile forces were distributed directly to the adjacent regions. A load of 500N was applied vertically at the first node from the lingual slope of the buccal cusp tip. This study analysed von Mises stress, 1 and 2 directional normal stress and Y and Z axis translation with FEM software Super SAPII Version 5.2 (Algor Interactive System Co.) and hardware 486 DX2 PC. The results were as :follows : 1. 1S model was slightly different with 1B model in stress distibution. 1S, 2B, 3B, 4B models showed similiar stress distribution. 2. 1S model and four B models showed similiar pattern in Y axis and Z axis translation. 3. 1S model and four B models showed the bending phenomenon in the translation. 4. As increasing of the width of the cavity, experimental group was similiar with the control group in stress distribution. 5. As increasing of the width of the cavity, experimental group was similiar with the control group in Y and Z axis tranlation.

• Korean Journal of Computational Design and Engineering
• /
• v.19 no.1
• /
• pp.41-49
• /
• 2014

For the reuse of the existing 3D block model of a ship, we analyze the hull modeling process using AVEVA Marine which is a representative CAD (Computer-Aided Design) system for the shipbuilding. In the AVEVA Marine environment where the design engineer makes 3D model on the 2D view that is so-called 2.5D, it cannot be possible to copy to reuse the block model just simply copying the 3D feature model itself like in the general mechanical CAD system or Smart Marine 3D which are on the basis of the 3D model representation. In this paper, we analyze the scheme file where the 3D model is defined in AVEVA Marine so that we develop the program for the block copy and the translation using this scheme file. It is significant that this program can be immediately available as a real-world application on the AVEVA Marine environment.