• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2023년도 제35회 한글 및 한국어 정보처리 학술대회
• /
• pp.352-357
• /
• 2023

자연어 추론은 두 문장(전제, 가설)간의 관계를 이해하고 추론하여 함의, 모순, 중립 세 가지 범주로 분류하며, 전제-가설-라벨(PHL) 데이터셋을 활용하여 자연어 추론 모델을 학습한다. 그러나, 새로운 도메인에 자연어 추론을 적용할 경우 학습 데이터가 존재하지 않거나 이를 구축하는 데 많은 시간과 자원이 필요하다는 문제가 있다. 본 논문에서는 자연어 추론을 위한 학습 데이터인 전제-가설-라벨 삼중항을 자동 생성하기 위해 [1]에서 제안한 문장 변환 규칙 대신에 거대 언어 모델과 Chain-of-Thought(CoT), Program-aided Language Models(PaL) 등의 프롬프팅(Prompting) 방법을 이용하여 전제-가설-라벨 삼중항을 자동으로 생성하는 방법을 제안한다. 실험 결과, CoT와 PaL 프롬프팅 방법으로 자동 생성된 데이터의 품질이 기존 규칙이나 기본 프롬프팅 방법보다 더 우수하였다.

• 한국농공학회논문집
• /
• 제52권5호
• /
• pp.1-9
• /
• 2010

Evapotranspiration (ET) is one of the basic components of the hydrologic cycle and is essential for estimating irrigation water requirements. In this study, artificial neural network (ANN) models for reference crop evapotranspiration ($ET_0$) estimation were developed on a monthly basis (May~October). The models were trained and tested for Suwon, Korea. Four climate factors, daily maximum temperature ($T_{max}$), daily minimum temperature ($T_{min}$), rainfall (R), and solar radiation (S) were used as the input parameters of the models. The target values of the models were calculated using Food and Agriculture Organization (FAO) Penman-Monteith equation. Future climate data were generated using LARS-WG (Long Ashton Research Station-Weather Generator), stochastic weather generator, based on HadCM3 (Hadley Centre Coupled Model, ver.3) A1B scenario. The evapotranspirations were 549.7 mm/yr in baseline period (1973-2008), 558.1 mm/yr in 2011-2030, 593.0 mm/yr in 2046-2065, and 641.1 mm/yr in 2080-2099. The results showed that the ANN models achieved good performances in estimating future reference crop evapotranspiration.

• 응용통계연구
• /
• 제33권6호
• /
• pp.713-722
• /
• 2020

본 논문에서는 금융시계열의 특징인 비대칭 변동성을 연구하고 있다. 멱변환을 동시에 고려한 멱변환-비대칭 GARCH 모형을 소개하고 있다. 변동성이 비정상인 모형을 다루고 있으며 오차항으로 표준정규분포와 더불어 표준화 t-분포도 고려하여 변동성 정상/비정상 조건을 제시하고 있다. 미국 주가 시계열인 다우지수 적용사례를 예시하였다.

• International Journal of Computer Science & Network Security
• /
• 제22권11호
• /
• pp.265-271
• /
• 2022

Social media is a form of communication based on the internet to share information through content and images. Their choice of profile images and type of image they post can be closely connected to their personality. The user posted images are designated as personality traits. The objective of this study is to predict five factor model personality dimensions from profile images by using deep learning and neural networks. Developed a deep learning framework-based neural network for personality prediction. The personality types of the Big Five Factor model can be quantified from user profile images. To measure the effectiveness, proposed two models using convolution Neural Networks to classify each personality of the user. Done performance analysis among two different models for efficiently predict personality traits from profile image. It was found that VGG-69 CNN models are best performing models for producing the classification accuracy of 91% to predict user personality traits.

• Journal of Advanced Research in Ocean Engineering
• /
• 제1권1호
• /
• pp.44-54
• /
• 2015

Breakwaters are the coastal structures constructed either perpendicular (shore connected) or parallel (detached) to the coast. The main function of breakwater is to create a tranquil medium on its leeside by reflecting the waves and also dissipating the wave energy arriving from seaside, resulting in ease of manoeuvrability to boats or ships to their berthing places. Different types of breakwaters are being used at present, such as rubble mound breakwater, vertical wall type breakwater and composite breakwater. The objective of this paper is to investigate reflection coefficients (Kr) and dissipation (loss) coefficients (Kl) for physical models of Quarter circle caisson breakwater of three different radii of 0.550 m, 0.575 m and 0.600 m with S/D ratio of 2.5 (S=spacing between perforations, D=diameter of perforations). The models were tested in the monochromatic wave flume of the department, for different incident wave heights (Hi), Wave periods (T) and water depths (d). It was observed that reflection coefficient increased with increase in the wave steepness (Hi/gT2) and decreased with increase in depth parameter (d/gT2) and hs/d (Height of structure including rubble base/depth of water). The loss coefficient decreased with increase in the wave steepness and increased with increase in depth parameter and hs/d.

• Communications for Statistical Applications and Methods
• /
• 제30권6호
• /
• pp.605-629
• /
• 2023

This paper proposes some diagnostics procedures for the skew-t linear regression model with censored response. The skew-t distribution is an attractive family of asymmetrical heavy-tailed densities that includes the normal, skew-normal and student's-t distributions as special cases. Inspired by the power and wide applicability of the EM-type algorithm, local and global influence analysis, based on the conditional expectation of the complete-data log-likelihood function are developed, following Zhu and Lee's approach. For the local influence analysis, four specific perturbation schemes are discussed. Two real data sets, from education and economics, which are right and left censoring, respectively, are analyzed in order to illustrate the usefulness of the proposed methodology.

• Journal of Astronomy and Space Sciences
• /
• 제26권4호
• /
• pp.425-438
• /
• 2009

ACE 위성에서 제공되는 1998년에서 2006년까지 9년간 관측된 양질의 1시간 평균 태양풍 매개변수를 이용하여 Burton et al.(1975)의 Dst 예측식을 재검토했다. 이 기간 동안 60개의 단순 발달형 자기폭풍을 골랐다. Burton et al.(1975)의 Dst 예측식의 에너지 공급항(Q)과 소멸 시간 ($\tau$)을 추정하기 위해 상기 자기폭풍 기간 동안 $Dst^*$와 $VS_s$, ${\Delta}Dst^*$와 $VS_s$ 그리고 ${\Delta}Dst^*$와 $Dst^*$의 상관관계를 구했다. 이 때 ACE 위성으로부터 지구까지 전파 시간(1시간)과 태양풍과 지구 자기권 사이의 지연 시간(0.5시간) 고려했다. 그 결과 $VB_s$ > 0.5mV/m일 때 $Q(nT/h)=-3.56VB_s$이였고, $VB_s\;{\leq}\;0.5mV/m$일 때는 Q(nT/h)=0으로 두었다(Burton et al., 1975) 그리고 $Dst^*$가 -175nT보다 작은 음의 값을 가질 때 $\tau(h)\;=\;0.060Dst^*\;+\;16.65$이고, $Dst^*$가 -175nT보다 큰 음의 값을 가질 때 $\tau(h)\;=\;6.15$로 추정됐다. 이 연구에서 얻은 Q와 $\tau$를 Burton et al.(1975)의 Dst 예측식에 대입하고 이를 이용하여 상기 60개 자기폭풍을 예측한 결과, 관측된 $Dst^*$와 예측된 $Dst^*$의 상관계수는 0.88이였다. 이를 다른 연구 결과와 비교하기 위해 Burton et al.(1975)과 O'Brien & McPherron(2000a)의 $Dst^*$ 예측 방법을 같은 자기폭풍에 적용한 결과, 관측된 $Dst^*$와 예측된 $Dst^*$의 상관계수는 각각 0.85였다. 이 연구는 기존 연구보다 다소 개선된 결과를 나타냈으며, 특히 $Dst^*\;{< \atop \sim}\;-200nT$의 강한 자기폭풍의 예측에 효과적이었다.

• 대한교통학회지
• /
• 제14권3호
• /
• pp.27-44
• /
• 1996

The purpose of the thesis is providing a new formulation for the transit assignment problem. The existing models dealing with the transit assignment problem don't consider the congestion effects due to the insufficient capacity of transit vehicles. Besides, these models don't provide solutions satisfying the Wardrop's user equilibrium conditions. The congestion effects are considered to be concentrated at the transit stops. For the transit lines, the waiting times at the transit stops are dependent on the passenger flows. The new model suggests the route section cost function analogous to the link performance function of the auto assignment to reflect the congestion effects in congested transit network. With the asymmetric cost function, the variational inequality programming is used to obtain the solutions satisfying Wardrop's condition. The diagonalization algorithm is introduced to solve this model. Finally, the results are compared with those of EMME/2.

• Computers and Concrete
• /
• 제5권4호
• /
• pp.401-416
• /
• 2008

The paper presents a methodology to model three-dimensional reinforced concrete members by means of embedded discontinuity elements based on the Continuum Strong Discontinuous Approach (CSDA). Mixture theory concepts are used to model reinforced concrete as a 3D composite material constituted of concrete with long fibers (rebars) bundles oriented in different directions embedded in it. The effects of the rebars are modeled by phenomenological constitutive models devised to reproduce the axial non-linear behavior, as well as the bond-slip and dowel action. The paper presents the constitutive models assumed for the components and the compatibility conditions chosen to constitute the composite. Numerical analyses of existing experimental reinforced concrete members are presented, illustrating the applicability of the proposed methodology.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권4호
• /
• pp.277-283
• /
• 2013

In this study, we propose a novel method for modeling dynamic hysteresis. Hysteresis is a widespread phenomenon that is observed in many physical systems. Many different models have been developed for representing a hysteretic system. Among them, the Duhem model is a classical nonlinear dynamic hysteresis model satisfying the properties of hysteresis. The purpose of this work is to develop a novel method that expresses the local dynamics of the Duhem model by a linear system model. Our approach utilizes a certain type of fuzzy system that is based on Takagi-Sugeno (T-S) fuzzy models. The proposed T-S fuzzy Duhem model is achieved by fuzzy blending of the linear system model. A simulated example applied to shape memory alloy actuators, which have typical hysteretic properties, illustrates the applicability of our proposed scheme.