• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.529-534
• /
• 2020

인공신경망 기반 기계번역(Neural Machine Translation, NMT)이란 딥러닝(Deep learning)을 이용하여 출발 언어의 문장을 도착 언어 문장으로 번역해주는 시스템을 일컫는다. NMT는 종단간 학습(end-to-end learning)을 이용하여 기존 기계번역 방법론의 성능을 앞지르며 기계번역의 주요 방법론으로 자리잡게 됐다. 이러한 발전에도 불구하고 여전히 개체(entity), 또는 전문 용어(terminological expressions)의 번역은 미해결 과제로 남아있다. 개체나 전문 용어는 대부분 명사로 구성되는데 문장 내 명사는 주체, 객체 등의 역할을 하는 중요한 요소이므로 이들의 정확한 번역이 문장 전체의 번역 성능 향상으로 이어질 수 있다. 따라서 본 논문에서는 지식그래프(Knowledge Graph)를 이용하여 심볼릭 지식을 NMT와 결합한 뉴럴심볼릭 방법론을 제안한다. 또한 지식그래프를 활용하여 NMT의 성능을 높인 선행 연구 방법론을 한영 기계번역에 이용할 수 있도록 구조를 설계한다.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.12 no.5
• /
• pp.590-600
• /
• 2009

In the simulation of underwater vehicles such as a submarine or a torpedo, various type of simulations like an engineering level simulation for predicting the performance precisely and an engagement level simulation for examining the effectiveness of a certain tactic is required. For this reason, a tactics manager which can change the behavior of a simulation model according to external tactics is needed. In this study the tactics manager supporting a script language and engine which can represent various tactics and can help users define external input tactics for the tactic manager easily is suggested. Python and Lua which are representative among script languages have been compared and analyzed from the viewpoint of a tactic manage, and the tactic manger using the script engines of those script languages was implemented. To demonstrate the effectiveness of the tactic manager, a target motion analysis simulation of the warfare between a submarine and a surface ship.

• Journal of the Korea Convergence Society
• /
• v.9 no.11
• /
• pp.23-28
• /
• 2018

Recent approaches to Relation Extraction methods mostly tend to be limited to mention level relation extractions. These types of methods, while featuring high performances, can only extract relations limited to a single sentence or so. The inability to extract these kinds of data is a terrible amount of information loss. To tackle this problem this paper presents an Augmented External Memory Neural Network model to enable Global Relation Extraction. the proposed model's Global relation extraction is done by first gathering and analyzing the mention level relation extraction by the Augmented External Memory. Additionally the proposed model shows high level of performances in korean due to the fact it can take the often omitted subjects and objectives into consideration.

• The Journal of Asian Finance, Economics and Business
• /
• v.9 no.4
• /
• pp.197-212
• /
• 2022

The study uses the audited financial statements of 26 Vietnamese commercial banks listed on the Ho Chi Minh City Stock Exchange (HOSE) and Hanoi Stock Exchange (HOSE) during the 2008-2018 period to estimate the system GMM model, which provides empirical evidence on the effect of the variables of customer deposit to total assets (DEPO) ratio, loan to assets (LTA) ratio, liquidity of commercial banks (LIQ), credit development (CRD) ratio, external funding (EFD) ratio, and credit loss provision (LLP) ratio on liquidity risk. The study confirms that commercial banks' internal factors play the most important role, and there is no empirical evidence on macro variables that affect liquidity risk. Finally, in accordance with the theoretical framework, the study uses an estimation method with the R language and the bootstrap methodology to give empirical proof of the nonlinear correlation and U-shaped graph between commercial bank size and liquidity risk. The importance of commercial bank size in absorbing and moderating the effects of liquidity shocks is demonstrated, however, excessive growth in commercial bank size would increase liquidity risk in commercial bank operations.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.11B
• /
• pp.1579-1587
• /
• 2010

In this paper, we design the PCS(Physical Coding Sublayer) transmitting and receiving module for 400/1000 Ethernet and verify the performance of it through logic simulation. In this work, we defined each function module and internal/external control signals and implemented them using HDL programming language. We also designed 64B/66B encoding/decoding, scrambling/descrambling including operation mode, detection of invalid frames, and multi-lane based distribution/arrangement. It was simulated using ModelSim and verified in terms of the operation and timing according to input data. The simulation result shows that all designed modules in 400/100G Ethernet are correctly performed.

• The Journal of the Korea Contents Association
• /
• v.18 no.7
• /
• pp.177-193
• /
• 2018

The business process management system that can integrate and manage a medical institution's processes has been increased importance to adapt to constantly changing medical environments and share information with various external medical institutions. The business process management system is an automated process tool that converts the graphic-based BPMN process model into a WS-BPEL, the execution language of the web service environment, and executes it through the process engine. However, the BPMN process model can be difficult to convert into WS-BPEL due to several ambiguities and structural inconsistencies. The process model may also contain structural anomalies that can lead to execution errors during process execution. In this paper, we present business process model formalization and structural anomaly verification techniques for facilitating integrated process management in medical institutions. Through the case study based on the IHE profile, we presented a formalized BPMN process model and verify the structural anomalies. We show the superiority of the proposed technique through comparative experiments with other related works.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.3
• /
• pp.91-100
• /
• 2020

Unlike the existing Visual Question Answering(VQA) problems, the new Visual Commonsense Reasoning(VCR) problems require deep common sense reasoning for answering questions: recognizing specific relationship between two objects in the image, presenting the rationale of the answer. In this paper, we propose a novel deep neural network model, KG_VCR, for VCR problems. In addition to make use of visual relations and contextual information between objects extracted from input data (images, natural language questions, and response lists), the KG_VCR also utilizes commonsense knowledge embedding extracted from an external knowledge base called ConceptNet. Specifically the proposed model employs a Graph Convolutional Neural Network(GCN) module to obtain commonsense knowledge embedding from the retrieved ConceptNet knowledge graph. By conducting a series of experiments with the VCR benchmark dataset, we show that the proposed KG_VCR model outperforms both the state of the art(SOTA) VQA model and the R2C VCR model.

• The KIPS Transactions:PartB
• /
• v.16B no.5
• /
• pp.427-434
• /
• 2009

With the rapid evolution of the Internet and mobile environments, text including spelling errors such as newly-coined words and abbreviated words are widely used. These spelling errors make it difficult to develop NLP (natural language processing) applications because they decrease the readability of texts. To resolve this problem, we propose a spelling error correction model using a spelling error correction dictionary and a newspaper corpus. The proposed model has the advantage that the cost of data construction are not high because it uses a newspaper corpus, which we can easily obtain, as a training corpus. In addition, the proposed model has an advantage that additional external modules such as a morphological analyzer and a word-spacing error correction system are not required because it uses a simple string matching method based on a correction dictionary. In the experiments with a newspaper corpus and a short message corpus collected from real mobile phones, the proposed model has been shown good performances (a miss-correction rate of 7.3%, a F1-measure of 97.3%, and a false positive rate of 1.1%) in the various evaluation measures.

• Water for future
• /
• v.28 no.1
• /
• pp.121-132
• /
• 1995

This paper presents a mathematical model which simulates the Cherepnov water lifter, that can lift water without the use of external energy such as electricity. A theoritical study was conducted to reveal the characteristics of the Cherepnov water lifter that was continuously operated with the aid of the siphon. The mathematical model wal composed of the continuity equation and energy equation, and was coded using FORTRAN language. In this study, the govering flow equation of the lifter were derived and the computer programs of the equations were worked out. The accuracy of the theoretical equations and their solutions was checked by laboratory experimentation. The mathematical model that simulate the Cherepnov water lifter is appeared to be good to predict the behavior of Cherepnov water lifter. Therefore, the mathematical model and the simulation method used in this study could be used in designing of the Cherepnov water lifter.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.11
• /
• pp.3991-4010
• /
• 2021

The Question Similarity Measurement of Chinese Crop Diseases and Insect Pests (QSM-CCD&IP) aims to judge the user's tendency to ask questions regarding input problems. The measurement is the basis of the Agricultural Knowledge Question and Answering (Q & A) system, information retrieval, and other tasks. However, the corpus and measurement methods available in this field have some deficiencies. In addition, error propagation may occur when the word boundary features and local context information are ignored when the general method embeds sentences. Hence, these factors make the task challenging. To solve the above problems and tackle the Question Similarity Measurement task in this work, a corpus on Chinese crop diseases and insect pests(CCDIP), which contains 13 categories, was established. Then, taking the CCDIP as the research object, this study proposes a Chinese agricultural text similarity matching model, namely, the AgrCQS. This model is based on mixed information extraction. Specifically, the hybrid embedding layer can enrich character information and improve the recognition ability of the model on the word boundary. The multi-scale local information can be extracted by multi-core convolutional neural network based on multi-weight (MM-CNN). The self-attention mechanism can enhance the fusion ability of the model on global information. In this research, the performance of the AgrCQS on the CCDIP is verified, and three benchmark datasets, namely, AFQMC, LCQMC, and BQ, are used. The accuracy rates are 93.92%, 74.42%, 86.35%, and 83.05%, respectively, which are higher than that of baseline systems without using any external knowledge. Additionally, the proposed method module can be extracted separately and applied to other models, thus providing reference for related research.