• Title/Summary/Keyword: Sequence-to-Sequence

Search Result 15,235, Processing Time 0.046 seconds

Korean Text Style Transfer Using Attention-based Sequence-to-Sequence Model (Attention-based Sequence-to-Sequence 모델을 이용한 한국어 어체 변환)

  • Hong, Taesuk;Xu, Guanghao;Ahn, Hwijeen;Kang, Sangwoo;Seo, Jungyun
    • Annual Conference on Human and Language Technology
    • /
    • 2018.10a
    • /
    • pp.567-569
    • /
    • 2018
  • 한국어의 경어체는 종결어미에 따라 구분하고, 서로 다른 경어체는 각각 고유한 경어 강도가 있다. 경어체 간의 어체 변환은 규칙기반으로 진행되어 왔다. 본 논문은 어체 변환을 위한 규칙 정의의 번거로움을 줄이고 어체 변환 데이터만을 사용한 심층 학습 기반의 어체 변환 방법을 제안한다. 본 연구는 '해요체-합쇼체' 쌍의 병렬 데이터를 이용하여 Attention-based Sequence-to-Sequence 모델을 바탕으로 한 어체 변환 모델을 학습하였다. 해당 모델을 학습하고 실험하였을 때, 정확도 91%의 우수한 성과를 얻을 수 있었다.

  • PDF

Improvement of Memory Efficiency for Alternative Sequence in Process Control System Described by SFC (SFC로 설계된 공정제어에서 선택시퀀스의 메모리효율향상)

  • You, Jeong-Bong
    • Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
    • /
    • v.24 no.5
    • /
    • pp.55-61
    • /
    • 2010
  • When we design the control system used Programmable Logic Controller(PLC) by Sequential Function Chart(SFC), if we use a SFC, it is easy to know the sequential flow of control, to maintenance the controller and to describe a program. We program a SFC by a unique sequence, an alternative sequence and a parallel sequence. If we program a SFC by a alternative sequence, the memory size of a alternative sequence must be larger than the memory size of a unique sequence. Therefore this thesis show an efficient method to reduce a memory size and we confirmed its feasibility through actual example.

Estimation of Physical Layer Scrambling Code Sequence of DVB-S2

  • Wu, Hao;Xie, Hui;Huang, Zhi-Tao;Zhou, Yi-Yu
    • ETRI Journal
    • /
    • v.36 no.2
    • /
    • pp.329-332
    • /
    • 2014
  • In this letter, the problem of estimating the physical layer (PL) scrambling code sequence of DVB-S2 is studied. We present the first ever scheme to estimate the scrambling sequence. The scheme is based on hypothesis testing. By analyzing the PL scrambling process, we construct a new sequence equivalent to the scrambling sequence. We then use hypothesis testing to estimate the new sequence. The threshold for the hypothesis testing is also discussed. The experiment results show that the performance of our estimation scheme can work even under high BER.

The Stacking Sequence Optimization of Stiffened Laminated Curved Panels with Different Loading and Stiffener Spacing

  • Kim Cheol;Yoon In-Se
    • Journal of Mechanical Science and Technology
    • /
    • v.20 no.10
    • /
    • pp.1541-1547
    • /
    • 2006
  • An efficient procedure to obtain the optimal stacking sequence and the minimum weight of stiffened laminated composite curved panels under several loading conditions and stiffener layouts has been developed based on the finite element method and the genetic algorithm that is powerful for the problem with integer variables. Often, designing composite laminates ends up with a stacking sequence optimization that may be formulated as an integer programming problem. This procedure is applied for a problem to find the stacking sequence having a maximum critical buckling load factor and the minimum weight. The object function in this case is the weight of a stiffened laminated composite shell. Three different types of stiffener layouts with different loading conditions are investigated to see how these parameters influence on the stacking sequence optimization of the panel and the stiffeners. It is noticed from the results that the optimal stacking sequence and lay-up angles vary depending on the types. of loading and stiffener spacing.

Comparison of Similarity to Digital Watermarking using Various Sequences (디지털 워터마킹을 위한 각종 시퀀스의 유사도 비교)

  • 송상주;박두순;김선형
    • Journal of the Korea Society of Computer and Information
    • /
    • v.6 no.4
    • /
    • pp.21-29
    • /
    • 2001
  • We show that image make converts of multiplex resolution using wavelet transform algorithm. To evaluate the robustness. we have measured comparison or similarity using various sequences which is inserted important coefficient of middle frequency band. The wavelet transform is advantage that it has a special quality of frequency domain and a special quality of spatial domain. Watermark is used pseudo random number, gaussian sequence, chaos sequence and sobel sequence. As result of experiments, it is to certify that The chaos sequence similarity is higher than other sequence. So the chaos sequence will be used for watermark sequence.

  • PDF

A New Galaxy Classification Scheme in the WISE Color-Luminosity Diagram

  • Lee, Gwang-Ho;Sohn, Jubee;Lee, Myung Gyoon
    • The Bulletin of The Korean Astronomical Society
    • /
    • v.38 no.2
    • /
    • pp.49.1-49.1
    • /
    • 2013
  • We present a new galaxy classification scheme in the Wide-field Infrared Survey Explorer (WISE) [$3.4{\mu}m$]-[$12{\mu}m$] color versus $12{\mu}m$ luminosity diagram. In this diagram, galaxies can be classified into three groups in different evolutionary stages. Late-type galaxies are distributed linearly along "MIR star-forming sequence" identified by Hwang et al. (2012). Some early-type galaxies show another sequence at [3.4]-[12] $(AB){\simeq}-2.0$, and we call this 'MIR blue sequence'. They are quiescent systems with old stellar population older than 10 Gyr. Between the MIR star-forming sequence and the MIR blue sequence, some early- and late-type galaxies are sparsely distributed, and we call these galaxies 'MIR green cloud galaxies'. Interestingly, both MIR blue sequence galaxies and MIR green cloud ones lie on the red sequence in the optical color-magnitude diagram. However, MIR green cloud galaxies have lower stellar masses and younger stellar populations (smaller $D_n4000$) than MIR blue sequence galaxies, suggesting that MIR green cloud galaxies are in the transition stage from MIR star-forming sequence galaxies to MIR blue sequence ones. We present differences in various galaxy properties between the three MIR classes using a multi-wavelength data, combined with the WISE and Sloan Digital Sky Survey Data Release 10, of local (0.03 < z < 0.07) galaxies.

  • PDF

Effects of Guanidination with Trypsin, Lys-C, or Glu-C Digestion on Mass Spectrometric Signal Intensity and Protein Sequence Coverage

  • Han, Hye-Sun;Nho, Seon-Ho;Lee, Ae-Ra;Kim, Jeong-Kwon
    • Bulletin of the Korean Chemical Society
    • /
    • v.31 no.6
    • /
    • pp.1527-1534
    • /
    • 2010
  • The conventional peptide modification process of guanidination, in which the amino groups of lysine residues are converted to guanidino groups using O-methylisourea to create more basic homoarginine residues, is often used to improve the signal intensity of lysine-containing peptides in matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). Here, we used three different protease enzymes (trypsin, Lys-C, and Glu-C) to evaluate the effects of guanidination on the MS signals of two enzymatically digested proteins. Horse heart myoglobin and bovine serum albumin were guanidinated either before or after digestion with trypsin, Lys-C, or Glu-C. The resulting peptides were subjected to MALDI-MS, and signal intensities and sequence coverage were systematically evaluated for each digest. Guanidination prior to Glu-C digestion improved sequence coverage for both proteins. For myoglobin, guanidination before enzymatic digestion with trypsin or Lys-C also enhanced sequence coverage, but guanidination after enzymatic digestion enhanced sequence coverage only with Lys-C. For albumin, guanidination either before or after Glu-C digestion increased sequence coverage, whereas pre- or post-digestion guanidination decreased sequence coverage with trypsin and Lys-C. The amino acid composition of a protein appears to be the major factor determining whether guanidination will enhance its MALDI-MS sequence coverage.

The Design of High Speed Processor for a Sequence Logic Control using FPGA (FPGA를 이용한 시퀀스 로직 제어용 고속 프로세서 설계)

  • Yang, Oh
    • The Transactions of the Korean Institute of Electrical Engineers A
    • /
    • v.48 no.12
    • /
    • pp.1554-1563
    • /
    • 1999
  • This paper presents the design of high speed processor for a sequence logic control using field programmable gate array(FPGA). The sequence logic controller is widely used for automating a variety of industrial plants. The FPGA designed by VHDL consists of program and data memory interface block, input and output block, instruction fetch and decoder block, register and ALU block, program counter block, debug control block respectively. Dedicated clock inputs in the FPGA were used for high speed execution, and also the program memory was separated from the data memory for high speed execution of the sequence instructions at 40 MHz clock. Therefore it was possible that sequence instructions could be operated at the same time during the instruction fetch cycle. In order to reduce the instruction decoding time and the interface time of the data memory interface, an instruction code size was implemented by 16 bits or 32 bits respectively. And the real time debug operation was implemented for easy debugging the designed processor. This FPGA was synthesized by pASIC 2 SpDE and Synplify-Lite synthesis tool of Quick Logic company. The final simulation for worst cases was successfully performed under a Verilog HDL simulation environment. And the FPGA programmed for an 84 pin PLCC package was applied to sequence control system with inputs and outputs of 256 points. The designed processor for the sequence logic was compared with the control system using the DSP(TM320C32-40MHz) and conventional PLC system. The designed processor for the sequence logic showed good performance.

  • PDF

Ship block assembly sequence planning considering productivity and welding deformation

  • Kang, Minseok;Seo, Jeongyeon;Chung, Hyun
    • International Journal of Naval Architecture and Ocean Engineering
    • /
    • v.10 no.4
    • /
    • pp.450-457
    • /
    • 2018
  • The determination of assembly sequence in general mechanical assemblies plays an important role in terms of manufacturing cost, duration and quality. In the production of ships and offshore plants, the consideration of productivity factors and welding deformation is crucial in determining the optimal assembly sequence. In shipbuilding and offshore industries, most assembly sequence planning has been done according to engineers' decisions based on extensive experience. This may result in error-prone planning and sub-optimal sequence, especially when dealing with unfamiliar block assemblies composed of dozens of parts. This paper presents an assembly sequence planning method for block assemblies. The proposed method basically considers geometric characteristics of blocks to determine feasible assembly sequences, as well as assembly process and productivity factors. Then the assembly sequence with minimal welding deformation is selected based on simplified welding distortion analysis. The method is validated using an asymmetric assembly model and the results indicate that it is capable of generating an optimal assembly sequence.

Automatic Conversion of English Pronunciation Using Sequence-to-Sequence Model (Sequence-to-Sequence Model을 이용한 영어 발음 기호 자동 변환)

  • Lee, Kong Joo;Choi, Yong Seok
    • KIPS Transactions on Software and Data Engineering
    • /
    • v.6 no.5
    • /
    • pp.267-278
    • /
    • 2017
  • As the same letter can be pronounced differently depending on word contexts, one should refer to a lexicon in order to pronounce a word correctly. Phonetic alphabets that lexicons adopt as well as pronunciations that lexicons describe for the same word can be different from lexicon to lexicon. In this paper, we use a sequence-to-sequence model that is widely used in deep learning research area in order to convert automatically from one pronunciation to another. The 12 seq2seq models are implemented based on pronunciation training data collected from 4 different lexicons. The exact accuracy of the models ranges from 74.5% to 89.6%. The aim of this study is the following two things. One is to comprehend a property of phonetic alphabets and pronunciations used in various lexicons. The other is to understand characteristics of seq2seq models by analyzing an error.