• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제14권1호
• /
• pp.1-7
• /
• 2014

This paper presents a novel method for simultaneously and automatically choosing the nonlinear structures of regressors or discriminant functions, as well as the number of terms to include in a rule-based regression model or pattern classifier. Variables are first partitioned into subsets each of which has a linguistic term (called a causal condition) associated with it; fuzzy sets are used to model the terms. Candidate interconnections (causal combinations) of either a term or its complement are formed, where the connecting word is AND which is modeled using the minimum operation. The data establishes which of the candidate causal combinations survive. A novel theoretical result leads to an exponential speedup in establishing this.

• Communications for Statistical Applications and Methods
• /
• 제31권2호
• /
• pp.191-202
• /
• 2024

The goal of this paper is to show how multivariate regression analysis with high-dimensional responses is facilitated by the response dimension reduction. Multivariate regression, characterized by multi-dimensional response variables, is increasingly prevalent across diverse fields such as repeated measures, longitudinal studies, and functional data analysis. One of the key challenges in analyzing such data is managing the response dimensions, which can complicate the analysis due to an exponential increase in the number of parameters. Although response dimension reduction methods are developed, there is no practically useful illustration for various types of data such as so-called large p-small n data. This paper aims to fill this gap by showcasing how response dimension reduction can enhance the analysis of high-dimensional response data, thereby providing significant assistance to statistical practitioners and contributing to advancements in multiple scientific domains.

• 응용통계연구
• /
• 제26권3호
• /
• pp.389-399
• /
• 2013

본 연구에서는 가우시안 과정회귀방법을 소개하고 시계열 마이크로어레이 유전자 발현데이터에 대해 가우시안 과정회귀를 적용한 사례를 보이고자한다. 가우시안 과정회귀를 적합하여 로그 주변우도함수 비를 이용한 유전자를 선별방법에 대한 모의실험을 통해 민감도, 특이도, 위발견율 등을 계산하여 선별방법으로의 활용성을 보였다. 실제 효모세포주기 데이터에 대해 제곱지수공분산함수를 고려한 가우시안 과정회귀를 적합하여 로그 주변우도함수 비를 이용하여 차변화된 유전자를 선별한 후, 선별된 유전자들에 대해 가우시안 모형기반 군집화를 하고 실루엣 값으로 군집유효성을 보였다.

• 한국물환경학회지
• /
• 제30권1호
• /
• pp.8-15
• /
• 2014

In this study, regression equation was analyzed to estimate non-point source (NPS) pollutant loads in orchard area. Many factors affecting the runoff of NPS pollutant as precipitation, storm duration time, antecedent dry weather period, total runoff density, average storm intensity and average runoff intensity were used as independent variables, NPS pollutant was used as a dependent variable to estimate multiple regression equation. Based on the real measurement data from 2008 to 2012, we performed correlation analysis among the environmental variables related to the rainfall NPS pollutant runoff. Significance test was confirmed that T-P ($R^2=0.89$) and BOD ($R^2=0.79$) showed the highest similarity with the estimated regression equations according to the NPS pollutant followed by SS and T-N with good similarity ($R^2$ >0.5). In the case of regression equation to estimate the NPS pollutant loads, regression equations of multiplied independent variables by exponential function and the logarithmic function model represented optimum with the experimented value.

• 생태와환경
• /
• 제57권2호
• /
• pp.111-122
• /
• 2024

Zooplankton biomass is essential for understanding the quantitative structure of lake food webs and for the functional assessment of biotic interactions. In this study, we aimed to propose a biomass (dry weight) estimation method using the body length of cyclopoid copepods. These copepods play an important role as omnivores in lake zooplankton communities and contribute significantly to biomass. We validated several previously proposed estimation equations against direct measurements and compared the suitability of prosomal length versus total length of copepods to suggest a more appropriate estimation equation. After comparing the regression analysis results of various candidate equations with the actual values measured on a microbalance-using the coefficient of variation, mean absolute error, and coefficient of determination-it was determined that the Total Length-DW exponential regression equation [W=0.7775×e^2.0183L; W (㎍), L (mm)] could be used to calculate biomass with higher accuracy. However, considering practical issues such as the morphological similarity between species and genera of copepods and the limitations of classifying copepodid stages, we derived a general regression equation for the pooled copepod community rather than a species-specific regression equation.

• 한국터널지하공간학회 논문집
• /
• 제10권3호
• /
• pp.233-243
• /
• 2008

터널 건설시 조사 및 설계 단계에서 예측한 터널의 거동은 지반의 지질 구조 및 재료 특성의 복잡함 등으로 인해 실제 시공 중의 거동과 일치하지 않는 경우가 많다. 따라서 시공 중에는 관찰과 계측을 통하여 지반과 지보재의 거동을 확인하고 위험이 예상될 경우 신속히 굴착 방법 및 보강 방법 등을 변경하므로써 터널의 안정성을 확보해야 한다. 그리고 이러한 목적을 위해서는 가급적 굴착 초기 단계에서 터널의 변형 거동을 파악하고 이로부터 최종적인 변위를 예측하는 과정이 필요하다. 본 연구에서는 NATM 방식에 의해 시공 중인 국내 두 곳의 도로터널 현장에서 수집한 변위 계측 data를 분석하여 굴착 초기 측정 내공변위량과 최종 내공변위량과의 관계를 통계적으로 유도하였다. 이 때 굴착 후 계측 전 까지 미측정 내공변위량을 추정하기 위하여 기존의 근사함수식을 수정하여 회귀분석하는 방법과 본 연구에서 제안하는 막장거리 1D(D:터널직경)까지의 계측 data를 이용하여 현장에서 용이하게 적용할 수 있는 간단한 선형 회귀분석을 이용하는 방법을 적용하였다. 최종적으로, 계측 전 변위량을 포함한 굴착 초기 내공변위와 최종 내공변위와의 관계를 통계적 방법을 통해 선형관계식으로 근사적으로 표현할 수 있었으며, 이 때 계측 전 변위량 추정방법으로 지수함수를 이용한 경우와 선형 회귀분석을 이용한 경우에 대해서 결과의 차이는 그다지 크지 않은 것으로 나타났다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1993년도 한국자동제어학술회의논문집(국내학술편); Seoul National University, Seoul; 20-22 Oct. 1993
• /
• pp.754-759
• /
• 1993

An adaptive model predictive control (AMPC) strategy using auto-regression moving-average (ARMA) models is presented. The characteristic features of this methodology are the small computer memory requirement, high computational speed, robustness, and easy handling of nonlinear and time varying MIMO systems. Since the process dynamic behaviors are expressed by ARMA models, the model parameter adaptation is simple and fast to converge. The recursive least square (RLS) method with exponential forgetting is used to trace the process model parameters assuming the process is slowly time varying. The control performance of the AMPC is verified by both comparative simulation and experimental studies on distillation column control.

• 한국지하수토양환경학회:학술대회논문집
• /
• 한국지하수토양환경학회 2003년도 총회 및 춘계학술발표회
• /
• pp.207-210
• /
• 2003

In this research analyzed the daily precipitation depths of 4 climatic stations in south eastern part of Korea. It aimed to present some indices for regionalization of the area. The items are frequency distributions of precipitation quantity and days and the longest days less than class limit set up by arithmetic differences. The regression analysis between class value and frequency show very good correlation coefficients better than 0.99 which are cubic equation for the precipitation, exponential equation for the precipitation days, and first degree equation for the longest day less than class limit.

• 터널과지하공간
• /
• 제13권2호
• /
• pp.108-116
• /
• 2003

백악기 경상계 안산암과 불국사 화강암류가 주로 분포하는 지반에서 시공된 터널의 내공변위 계측자료를 분석하였다. 터널 주변 암반을 RMR법에 의한 다섯 가지 암반등급으로 구분하고 각 등급에 포함된 계측자료들을 통계처리하여 암반등급별 내공변위의 회귀분석을 실시하였다. 연구 결과. 로그함수보다는 지수함수의 상관계수 가 더 크며, 연약한 암반등급일수록 내공변위의 크기와 표준편차가 크게 나타났다. 또한, 최종내공변위에 대한 최대변위속도 및 초기내공변위의 관계를 도출하였으며, 이 중에서 최종내공변위와 최대변위속도의 상관계수는 0.87로 나타나 이들은 비교적 높은 상관성을 가지는 것으로 확인되었다

• 품질경영학회지
• /
• 제12권2호
• /
• pp.25-32
• /
• 1984

The purpose of this paper is to compare several forecasting methods for the case of the cement product by the analysis of the forecasting data and by the study of major forecasting methods, which are the Trend Projection, Exponential Smoothing, and Multiple Regression Analysis. As a result, it is thought that the Multiple Regression Analysis is the optimal model for the case of the cement product. In addition, it is important to consider the future circumstances for forecasting, and to improve the level of the forecasting results through the precise analysis of the collected data.