• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2001년도 추계학술발표논문집 (상)
• /
• pp.535-538
• /
• 2001

Since 1990s, many literatures have shown that connectionist models, such as back propagation, recurrent network, and RBF (Radial Basis Function) outperform the traditional models, MA (Moving Average), AR (Auto Regressive), and ARIMA (Auto Regressive Integrated Moving Average) in time series prediction. Neural based approaches to time series prediction require the enough length of historical measurements to generate the enough number of training patterns. The more training patterns, the better the generalization of MLP is. The researches about the schemes of generating artificial training patterns and adding to the original ones have been progressed and gave me the motivation of developing VTG schemes in 1996. Virtual term is an estimated measurement, X(t+0.5) between X(t) and X(t+1), while the given measurements in the series are called actual terms. VTG (Virtual Tern Generation) is the process of estimating of X(t+0.5), and VTG schemes are the techniques for the estimation of virtual terms. In this paper, the alternative VTG schemes to the VTG schemes proposed in 1996 will be proposed and applied to multivariate time series prediction. The VTG schemes proposed in 1996 are called deterministic VTG schemes, while the alternative ones are called stochastic VTG schemes in this paper.

• 한국초등과학교육학회지:초등과학교육
• /
• 제30권3호
• /
• pp.353-366
• /
• 2011

An alternative vision for science inquiry that appears to be important and challenging is model-based inquiry in which students generate, evaluate and revise their explanatory model. Pre-service teachers should be given opportunities to develop and use their mechanistic explanatory models in order to participate in the practice of science and to have a sound understanding of science. With this view, this study described a case of pre-service elementary teachers' scientific modeling in magnetism. The aims of this study were to explore difficulties preservice elementary teachers encountered while they engaged in a model-based inquiry, and to examine how their understandings of the nature of scientific models changed after the model-based inquiry. The data analysis revealed that the pre-service teachers had difficulties in drawing and writing their own thinking because they had little experience of expressing their own science ideas. When asked to predict what would happen, they could not understand what it meant to make a prediction "based on their model". They did not know how to use or consider their model in making a prediction. At the end of the model-based inquiry they reached a final consensus of a best model. However, they were very anxious about whether the model was the "correct" answer. With respect to the nature of scientific models, almost all of the pre-service teachers initially viewed models only as a communication tool among scientists or students and teachers to help understand others' ideas. After the model-based inquiry, however, many of them understood that they could create, test, and revise their "own" models "by themselves". They also realized the key aspects of scientific models that a model can be changed as evidence is accumulated and a model is a knowledge production tool as well as a communication tool. The results indicated that pre-service elementary teachers' understandings of the nature of scientific models and their previous school science experiences could affect their performance on a model-based inquiry, and their experience of scientific modeling could help them enhance their understandings of the nature of scientific models.

• 지적과 국토정보
• /
• 제48권2호
• /
• pp.197-211
• /
• 2018

본 연구에서는 건강에 대한 공간적 연구에서 통상적으로 사용되는 센서스에 기반한 지역 박탈지수의 대안으로 지역 주택가격이 사용될 수 있는지 평가하였다. 평가를 위해 개인을 1수준으로, 수도권의 보건소 구역을 2수준으로 하는 다수준 로지스틱 모델이 추정되었다. 다수준 모델에는 개인의 점심식사후 칫솔질과 치간실 사용을 설명하기 위한 개인수준의 변수들과 보건소 구역을 대표하는 사회적 박탈지수 및 지역주택가격 수준이 포함되었다. 추정된 모델들의 설명력은 Akaike Information Criterion (AIC)와 Bayesian Information Criterion (BIC)를 이용하여 평가되었다. 모델의 추정결과는 사회적 박탈지수 및 지역 주택가격이 모두 개인의 치아관리 행동을 설명하는 데 기여하나 지역 주택가격을 사용한 모델의 AIC 및 BIC가 통상적인 센서스 기반 지역 박탈지수를 사용한 경우 보다 낮은 것을 보여 주었다. 본 연구결과는 센서스에 기반한 박탈지수를 생성하는 데 사용된 센서스 변수가 시점의 차이 등의 이유로 적절하지 않을 경우 지역 주택가격이 지역의 사회경제적 수준을 대표하기 위해 대안적으로 사용될 수 있음을 보여준다.

• 한국유기농업학회지
• /
• 제25권3호
• /
• pp.555-567
• /
• 2017

This study suggests the yield forecast model for chilli pepper using artificial neural network. For this, we select the most suitable network models for chilli pepper's yield and compare the predictive power with adaptive expectation model and panel model. The results show that the predictive power of artificial neural network with 5 weather input variables (temperature, precipitation, temperature range, humidity, sunshine amount) is higher than the alternative models. Implications for forecasting of yields are suggested at the end of this study.

• Communications for Statistical Applications and Methods
• /
• 제19권1호
• /
• pp.135-145
• /
• 2012

It is well known that the maximum likelihood estimator(MLE) in normal mixture models with unequal variances does not fall in the interior of the parameter space. Recently, a doubly smoothed maximum likelihood estimator(DS-MLE) (Seo and Lindsay, 2010) was proposed as a general alternative to the ordinary maximum likelihood estimator. Although this method gives a natural modification to the ordinary MLE, its computation is cumbersome due to intractable integrations. In this paper, we derive an EM algorithm for the DS-MLE under normal mixture models and propose a fast computational tool using a local quadratic approximation. The accuracy and speed of the proposed method is then presented via some numerical studies.

• Korean Journal of Hydrosciences
• /
• 제4권
• /
• pp.33-49
• /
• 1993

The Purpose of this study is to classify existing hysteresis models and to discuss a possibility of a new type of the hysteresis model. The existing hysteresis models are classified into three types: the interpolation model, the scaling model and the domain model, of which only domain model is to simulate hysteresis curves based on the theoretical approach, It is useful to develop a hysteresis model that requires only one branch of hysteresis curves for the model calibration because obtaining hysteresis curves by experiments is expensive and time-concept by many investigators, however their models are not successful to accurately simulate real data of Rubicon Sandy Loam and Dune Sand. There is a possibility that a new model is based on the dependent domain concept considering the weighting factor, $P_a$($\theta$), which accounts for the pore blockage effect against air entry. Conclusively, a new model where the weighting factor $P_a$($\theta$) in Model III-1 (Mualem, 1984) reduces to a known variable through an appropriate method is an alternative model which required only one branch of main curves for the model calibration.

• Wind and Structures
• /
• 제2권4호
• /
• pp.305-323
• /
• 1999

Accurate turbulence modeling is an essential prerequisite for the use of Computational Fluid Dynamics (CFD) in Wind Engineering. At present the most popular turbulence model for general engineering flow problems is the ${\kappa}-{\varepsilon}$ model. Models such as this are based on the isotropic eddy viscosity concept and have well documented shortcomings (Murakami et al. 1993) for flows encountered in Wind Engineering. This paper presents an objective assessment of several available alternative models. The CFD results for the flow around a full-scale (6 m) three-dimensional surface mounted cube in an atmospheric boundary layer are compared with recently obtained data. Cube orientations normal and skewed at $45^{\circ}$ to the incident wind have been analysed at Reynolds at Reynolds number of greater than $10^6$. In addition to turbulence modeling other aspects of the CFD procedure are analysed and their effects are discussed.

• 응용통계연구
• /
• 제13권2호
• /
• pp.437-445
• /
• 2000

국내 주가시계열을 분석하기 위해 기존의 비선형시계열모형인 분계점을 가진 자기외귀모형(TAR)과 일반화 이분산자기회귀모형(GARCH)을 비교 분석한 후, 이 두가지 모형을 결합시킨 새로운 모형 TAT-GARCH모형을 제안하였다. 이 모형은 그 자체로도 이론적인 관삼의 대상이 되어 연관된 모수추정 기법을 제시하였고 국내 개별 주가시계열 자료의 분석에 있어서 제안된 모형이 기존의 모형들 보다 상대적으로 더 좋은 예측치를 제공할 수 있음을 특정 9개 회사의 주가분석을 통해 알아보았다.

• 한국토양비료학회지
• /
• 제42권Spc호
• /
• pp.8-13
• /
• 2009

심층토 내에서의 흐름과 이동과정을 포함하는 연구와 관리 측면에서 수학모델에 대한 관심도가 점점 증가하고 있다. 로제타는 토성이나 용적밀도 자료와 같은 대체 토양자료로부터 불포화수리특성에 대한 자료를 추정하는 프로그램이다. 이와 같은 형태의 모델은 애초 기본 토양자료를 수리특성자료로 전환하기 시작한 이래 대체 측정수단으로서 PTF라 불리워졌다. 이러한 기능은 유사-실험모델을 사용하여 예측한 자료를 근간으로 하여 직간접적으로 토양수분을 추정할 수 있다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2006년도 정기 학술대회 논문집
• /
• pp.1048-1055
• /
• 2006

Two ways can be used for dynamic condensation of viscously damped structural models. One is reducing the model in physical space at first and then transferring it to state space. The other is ,condensing the model in state space directly. Two iterative schemes for each way are given respectively. Hence four iterative schemes for dynamic condensation of nonclassically damped models are discussed in this paper. A high building with a tuned mass damper is applied to show the efficiency of these schemes.