• The Journal of the Acoustical Society of Korea
• /
• v.22 no.7
• /
• pp.528-538
• /
• 2003

A SCCRS (Speech and Character Combined Recognition System) is developed for working on mobile devices such as PDA (Personal Digital Assistants). In SCCRS, the feature extraction is separately carried out for speech and for hand-written character, but the recognition is performed in a common engine. The recognition engine employs essentially CHMM (Continuous Hidden Markov Model), which consists of variable parameter topology in order to minimize the number of model parameters and to reduce recognition time. For generating contort independent variable parameter model, we propose the SSMS(Successive State and Mixture Splitting), which gives appropriate numbers of mixture and of states through splitting in mixture domain and in time domain. The recognition results show that the proposed SSMS method can reduce the total number of GOPDD (Gaussian Output Probability Density Distribution) up to 40.0% compared to the conventional method with fixed parameter model, at the same recognition performance in speech recognition system.

• Computational Structural Engineering
• /
• v.9 no.1
• /
• pp.65-76
• /
• 1996

The initiation and growth of microcracks or microvoids inside concrete results in the progressive degradation of concrete. This damage processing along processing along with plastic deformation is main cause of nonlinear behavior of concrete. In this study, a continuum damage model of concrete is developed for the analysis of the nonlinear behavior of concrete due to damage and elasto-plastic deformation. Anisotropic damage tensor is used to describe the anisotropy of concrete and hypothesis of equivalent elastic energy is used to define the effective elastic tensor. The damage model including the damage evolution law and constitutive equation is derived with damage variable and damage surface which is defined by damage energy release rate by using the Helmholtz free energy and dissipation potential based on the thermodynamic principles. By adopting a typical plasticity model of concrete, plasticity of concrete is included to this model. Afinite element analysis program implemented with this model was developed and finite element analysis was performed for the analyses of concrete subjected to uniaxial and biaxial loadings. Comparison of the results of analysis with those of experiments and other models shows that the model successfully predicts the nonlinear behavior of concrete.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.15 no.2
• /
• pp.116-126
• /
• 2003

In this study, a method that leads to make a simple decision on important parameters in analysis of wave field in permeable rubble-mound, block-mound breakwater, such as penetration velocity of incident waves and resistance coefficient, is introduced. A model that could analyze wave field of permeable breakwater in harbor, by applying these methods and arbitrary transmission coefficient boundary condition to a time-dependent mild-slope equation, was introduced. The verification of the model was done by carrying out 2-D physical model test on permeable breakwater, measuring the change in water surface elevation, comparing the computation result with time series, and comparing the result gained from the 3-D physical model test on permeable block-mound breakwater in an field harbor with the computation result in terms of regional wave height ratio in a harbor.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.35 no.10
• /
• pp.1161-1170
• /
• 2011

This paper presents a solution to the inverse problem for the service boundary conditions of thermal-flow and structure analysis in a catalyst substrate. The exhaust-gas purification efficiency of a catalyst substrate is influenced by the shape parameter, catalyst ingredients and so on and is estimated by the thermal flow uniformity. The formulations of the inverse problem of obtaining the thermal-flow parameters (inlet temperature, velocity, heat of reaction, convective heat-transfer coefficient) and the direct problem of estimating from a given outlet temperature distribution are described. An experiment was designed and the response-surface optimization technique was used to solve the proposed inverse problem. The temperature distribution of the catalyst substrate was obtained by thermal-flow analysis for the predicted thermal-flow parameters. The thermal stress and durability assessments for the catalyst substrate were performed on the basis of this temperature distribution. The efficiency and accuracy of the inverse approach have been demonstrated through the achievement of good agreement between the thermal-flow response surface model and the results of experimental vehicle tests.

• Smart Media Journal
• /
• v.13 no.6
• /
• pp.24-34
• /
• 2024

If the future can be predicted from network traffic data, which is a time series, it can achieve effects such as efficient resource allocation, prevention of malicious attacks, and energy saving. Many models based on statistical and deep learning techniques have been proposed, and most of these studies have focused on improving model structures and learning algorithms. Another approach to improving the prediction performance of the model is to obtain a good-quality data. With the aim of obtaining a good-quality data, this paper applies a dense sampling technique that augments time series data to the application of network traffic prediction and analyzes the performance improvement. As a dataset, UNSW-NB15, which is widely used for network traffic analysis, is used. Performance is analyzed using RMSE, MAE, and MAPE. To increase the objectivity of performance measurement, experiment is performed independently 10 times and the performance of existing sparse sampling and dense sampling is compared as a box plot. As a result of comparing the performance by changing the window size and the horizon factor, dense sampling consistently showed a better performance.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.150-152
• /
• 2022

In this work, we propose a model that considers the behavior and synaptic plasticity of sensory neurons based on Liquid Time-constant Network (LTC). The neuron connection structure was experimented with four types: the increasing number of neurons, the decreasing number, the decreasing number, and the decreasing number. In this study, we experimented using a time series prediction dataset to see if the performance of the changed model improved compared to LTC. Experimental results show that the application of modeling of sensory neurons does not always bring about performance improvements, but improves performance through proper selection of learning rules depending on the type of dataset. In addition, the connective structure of neurons showed improved performance when it was less than four layers.

• Journal of the Korean Society for Advanced Composite Structures
• /
• v.3 no.1
• /
• pp.1-7
• /
• 2012

This study investigates free vibration characteristics of laminated composite structures with different embedded delamination sizes and locations using the solid element. The three-dimensional finite element (FE) delamination model described in this paper, in comparison with the conventional approaches, is more attractive not only because it shows better accuracy but also it shows the entire mode shape. The FE model using ABAQUS is used for studying free vibrations of laminates containing an various embedded delamination. The numerical results obtained are in good agreement with those reported by other investigators. In particular, new results reported in this paper are focused on the significant effects of the local vibration mode for various parameters, such as size of delamination, aspect ratio, and location of delamination.

• 한국해양학회지
• /
• v.27 no.1
• /
• pp.35-45
• /
• 1992

This paper is concerned with the development and application of a stochastic model for predicting E1 nino phenomenon. For this, first a general criterion for determining E1 nino phenomenon, including period and strength, which is based on partial sum of monthly sea surface temperatures (SST) anomalies, is proposed, Secondly, the annual fluctuations, periodicity and dependence of monthly mean of equatorial Pacific SST during the period 1951-1990 are analyzed. Based on these, time series nonlinear regression model for the prediction of SST have been derived. A statistical procedure for using the model to predict the SST have been derived. A statistical procedure for using the model to predict the SST level is also proposed.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.151-154
• /
• 2018

본 논문에서는 CNN Seq2Seq 구조를 이용해 한국어 대화 시스템을 개발하였다. 기존 Seq2Seq는 RNN 혹은 그 변형 네트워크에 데이터를 입력하고, 입력이 완료된 후의 은닉 층의 embedding에 기반해 출력열을 생성한다. 우리는 CNN Seq2Seq로 입력된 발화에 대해 출력 발화를 생성하는 대화 모델을 학습하였고, 그 성능을 측정하였다. CNN에 대해서는 약 12만 발화 쌍을 이용하여 학습하고 1만 발화 쌍으로 실험하였다. 평가 결과 제안 모델이 기존의 RNN 기반 모델에 비해 우수한 결과를 보였다.

• Journal of KIISE:Software and Applications
• /
• v.28 no.6
• /
• pp.429-438
• /
• 2001

정보통신 구조의 확산과 함께 전산시스템에 대한 침입과 피해가 증가되고 있으며 침입탐지 시스템에 대한 관심과 연구가 늘어나고 있다. 본 논문에서는 은닉 마르코프 모델(HMM)을 이용하여 사용자의 정상행위에서 생성된 이벤트ID 정보를 모델링한 후 사용자의 비정상행위를 탐지하는 침입탐지 시스템을 제안한다. 전처리를 거친 이벤트ID열은 전방향-역방향 절차와 Baum-Welch 재추정식을 이용하여 정상행위로 구축된다. 판정은 전방향 절차를 이용해서 판정하려는 열이 정상행위로부터 생성되었을 확률을 계산하며, 이 값을 임계값과 비교함으로써 수행된다. 실험을 통해 침입탐지를 위한 최적의 HMM 매개변수를 결정하고 사용자 구분이 없는 단일모델링, 사용자별 모델링, 사용자 그룹별 모델링 방식을 비교하여 정상행위 모델링 성능을 평가하였다. 실험결과 제안한 시스템이 발생한 침입을 적절히 탐지함을 확인할 수 있었지만, 신뢰도 높은 침입탐지 시스템의 구축을 위해서는 보다 정교한 모델의 클러스터링이 필요함을 알 수 있었다.