• 인터넷정보학회논문지
• /
• 제15권3호
• /
• pp.109-116
• /
• 2014

본 논문에서는 분류 성능을 향상하기 위해서 Takagi-Sugeno(T-S) 퍼지 모델 기반의 가중 퍼지소속함수 기반 신경망(Neural Network with Weighted Fuzzy Membership Functions; NEWFM)을 이용한 새로운 인스턴스 선택을 제안하였다. 제안하는 인스턴스 선택은 T-S 퍼지 모델에서의 가중 평균 역퍼지화와 통계학에서 사용하는 정규분포의 신뢰구간과 같은 구간 선택을 이용하여 인스턴스를 선택하였다. 제안하는 인스턴스 선택의 분류 성능을 평가하기 위해서 인스턴스 사용 전/후에 따라서 분류 성능을 비교하였다. 인스턴스 사용 전/후에 따른 분류 성능은 각각 77.33%, 78.19%로 나타났다. 또한 인스턴스 사용 전/후에 따른 분류 성능 간에 차이점을 보여주기 위해서 통계학에서 사용하는 맥니마 검정을 사용하였다. 맥니마 검정의 결과로 유의 확률이 0.05보다 적게 나오므로 인스턴스 선택의 분류 성능이 인스턴스 선택을 하지 않는 경우의 분류 성능보다 우수함을 확인 할 수가 있었다.

• Journal of Communications and Networks
• /
• 제6권3호
• /
• pp.269-279
• /
• 2004

A new online multi-layer integrated routing (MLIR) scheme that combines IP (electrical) layer routing with WDM (optical) layer routing is investigated. It is a highly efficient and cost-effective routing scheme viable for the next generation integrated optical Internet. A new simplified weighted graph model for the integrated optical Internet consisted of optical routers with multi-granularity optical-electrical hybrid switching capability is firstly proposed. Then, based on the proposed graph model, we develop an online integrated routing scheme called differentiated weighted fair algorithm (DWFA) employing adaptive admission control (routing) strategies with the motivation of service/bandwidth differentiation, which can jointly solve multi-layer routing problem by simply applying the minimal weighted path computation algorithm. The major objective of DWFA is fourfold: 1) Quality of service (QoS) routing for traffic requests with various priorities; 2) blocking fairness for traffic requests with various bandwidth granularities; 3) adaptive routing according to the policy parameters from service provider; 4) lower computational complexity. Simulation results show that DWFA performs better than traditional overlay routing schemes such as optical-first-routing (OFR) and electrical-first-routing (EFR), in terms of traffic blocking ratio, traffic blocking fairness, average traffic logical hop counts, and global network resource utilization. It has been proved that the DWFA is a simple, comprehensive, and practical scheme of integrated routing in optical Internet for service providers.

• Journal of the Korean Data and Information Science Society
• /
• 제8권2호
• /
• pp.153-161
• /
• 1997

본 연구에서는 우리나라 6대 생명보험회사의 모든 남자의 개인보험 계약자를 대상으로 조사한 1988년부터 1992년까지 25,000,000건의 자료를 가중이동평균모형을 이용한 보간법과 곰페 르츠모형을 이용한 보외법을 적용하여 경험생명표를 작성하였다. 적합성 및 평활성 기준에서 가중이동평균모형의 차분수와 항수, 곰페르츠모형의 모수 및 두 모형이 접촉하는 접점의 연령을 산출하였다. 특히 연립방정식을 이용하여 곰페르츠 모형의 모수를 폐쇄형으로 추정하는 방법을 제안하였다.

• 대한산업공학회지
• /
• 제32권4호
• /
• pp.279-290
• /
• 2006

This paper proposes a model to assess the maturity level of Six Sigma activities. We classify the maturity level into 5 stages: initial, forming, storming, performing and mature stage. To evaluate the maturity level, 10 categories of Six Sigma with 3 factors each are identified: management leadership, belt system, expert training, establishing execution system, compensation, organization, corporate culture, customer focus, project selection, and management of project results. Scoring 277 items in total, the value of each factor is evaluated by weighted average of those items. Maturity level is appraised by rating the sum of scores of 10 categories that are obtained by summing up the values of its 3 factors. Values of weights and criteria of rating maturity levels are determined by analyzing 90 companies and Six Sigma exper's opinion. This study also shows the actual appraisal results of some companies.

• 한국환경과학회지
• /
• 제22권3호
• /
• pp.303-308
• /
• 2013

This paper proposed a new method for estimating missing values in time series rainfall data. The proposed method integrated the two most widely used estimation methods, general linear model(GLM) and ordinary kriging(OK), by taking a weighted average of covariance matrices derived from each of the two methods. The proposed method was cross-validated using daily rainfall data at thirteen rain gauges in the Hyeong-san River basin. The goodness-of-fit of the proposed method was higher than those of GLM and OK, which can be attributed to the weighting algorithm that was designed to minimize errors caused by violations of assumptions of the two existing methods. This result suggests that the proposed method is more accurate in missing values in time series rainfall data, especially in a region where the assumptions of existing methods are not met, i.e., rainfall varies by season and topography is heterogeneous.

• 대한기계학회논문집B
• /
• 제32권4호
• /
• pp.317-326
• /
• 2008

The main purpose of the present study is to perform shape optimizations of transonic compressor blade in order to enhance its performance. In this study, the Latin hypercube sampling of design of experiments and the weighted average surrogate model with the help of a gradient based optimization algorithm are used within design space by the lower and upper limits of each design variable and for finding optimum designs, respectively. 3-D Reynolds-averaged Navier-Stokes solver is used to evaluate the objective functions of adiabatic efficiency and pressure ratio. Six variables from lean and airfoil thickness profile are selected as design variables. The results show that the adiabatic efficiency is enhanced by 1.43% by efficiency optimization while the pressure ratio is increased very small, and pressure ratio is increased by 0.24% by pressure ratio optimization.

• 한국수학교육학회지시리즈B:순수및응용수학
• /
• 제16권1호
• /
• pp.133-145
• /
• 2009

The purpose of this paper is to extend and globalize the Walrasian evolutionary cobweb model in an independent single local market of Brock and Hommes ([3]), to the case of the global market evolution over an infinite chain of many local markets interacting each other through a diffusion of prices between them. In the case of decreasing demands and increasing supplies with a weighted average of rational and naive predictors, we investigate, via the methods of Lattice Dynamical System, the spatial-temporal behaviors of global market dynamics and show that some kind of bounded dynamics of global market do exist and can be controlled by using the parameters in the model.

• 대한전기학회논문지
• /
• 제35권5호
• /
• pp.193-198
• /
• 1986

In this paper, 1-168 hours ahead load prediction algorithm is developed for power system economic weekly operation. Total load is composed of three components, which are base load, week load and weather-sensitive load. Base load and week load are predicted by moving average and exponential smoothing method, respectively. The days of moving average and smoothing constant are optimally determined. Weather-sensitive load is modeled by linear form. The paramiters of weather load model are estimated by exponentially weighted recursive least square method. The load prediction of special day is very tedious, difficult and remains many problems which should be improved. Test results are given for the day of different types using the actual load data of KEPCO.

• 환경영향평가
• /
• 제18권1호
• /
• pp.31-39
• /
• 2009

A lot of data which are used in environment analysis of air pollution have characteristics that are distributed continuously in space. In this point, the collected data value such as precipitation, temperature, altitude, pollution density, PM10 have spatial aspect. When geostatistical data analysis are needed, acquisition of the value in every point is the best way, however, it is impossible because of the costs and time. Therefore, it is necessary to estimate the unknown values at unsampled locations based on observations. In this study, spatial interpolation method such as local trend surface model, IDW(inverse distance weighted), RBF(radial basis function), Kriging were applied to PM10 annual average concentration of Seoul in 2005 and the accuracy was evaluated. For evaluation of interpolation accuracy, range of estimated value, RMSE, average error were analyzed with observation data. The Kriging and RBF methods had the higher accuracy than others.

• Communications for Statistical Applications and Methods
• /
• 제31권1호
• /
• pp.87-104
• /
• 2024

Most functional magnetic resonance imaging (fMRI) studies in resting state have assumed that the functional connectivity (FC) between time series from distinct brain regions is constant. However, increased interest has recently been in quantifying possible dynamic changes in FC during fMRI experiments. FC study may provide insight into the fundamental workings of brain networks to brain activity. In this work, we focus on the specific problem of estimating the dynamic behavior of pairwise correlations between time courses extracted from two different brain regions. We compare the sliding-window techniques such as moving average (MA) and exponentially weighted moving average (EWMA), dynamic causality with vector autoregressive (VAR) model, dynamic conditional correlation (DCC) based on volatility, and the proposed alternative methods to use differencing and recursive residuals. We investigate the properties of those techniques in a series of simulation studies. We also provide an application with major depressive disorder (MDD) patient fMRI data to demonstrate studying dynamic correlations.