• 산업경영시스템학회지
• /
• 제22권49호
• /
• pp.99-113
• /
• 1999

This paper describes a study which was undertaken in Ulsan City. It attempted to develop stepwise procedures that would aid Ulsan City in making decision of primary importance: what is the optimal site location for establishing a new Industrial Technology Research Park for public development\ulcorner The presented modeling procedures are an adaption of a number of exiting methods for the evaluation of industrial site potential. The procedures to determine the best site location can be divided into three phases : (i)defining the information necessary to compare potential sites, (ii) collecting the information for each site and (iii)evaluation potential sites utilizing the location model.

• Journal of the Korean Data and Information Science Society
• /
• 제23권3호
• /
• pp.457-465
• /
• 2012

본 연구의 목적은 한국대학교육협의회 대학정보공시센터의 '대학알리미'에서 주요 대학지표들을 분석하고, 지표들 간의 연관성과 통계적 모형을 추정하는데 있다. 먼저 상관계수에 대한 통계적 검정을 이용하여 변수들 간의 통계적으로 유의한 상관성을 추정하고, 이들 주요 지표들의 모형을 추정하기 위해서 회귀분석 방법의 변수선택 방법을 이용하여 회귀 방정식을 추정하여 변수들 간의 연관성을 조사하였다. 변수선택의 판정기준에 따른 방법으로 전진선택법과 후진제거법, 단계별 회귀방법을 사용하였다.

• Journal of Periodontal and Implant Science
• /
• 제50권2호
• /
• pp.68-73
• /
• 2020

Pain management is an important part of dental practice, and dentists frequently prescribe analgesics to improve clinical outcomes. Dentists should be aware of the pharmacological characteristics of the analgesics commonly used in dentistry and should choose appropriate analgesics to treat and prevent pain associated with inflammation or surgery. In this article, we review the potential benefits and risks of the analgesics frequently used in dental practice and provide a stepwise approach for pain management.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제12권2호
• /
• pp.59-76
• /
• 2023

코로나 시대 이후 아파트 가격 상승은 비상식적이었다. 이러한 불확실한 부동산 시장에서 가격 예측 연구는 매우 중요하다. 본 논문에서는 다양한 부동산 사이트에서 자료 수집 및 크롤링을 통해 2015년부터 2020년까지 87만개의 방대한 데이터셋을 구축하고 다양한 아파트 정보와 경제지표 등 가능한 많은 변수를 모은 뒤 미래 아파트 매매실거래가격을 예측하는 모델을 만든다. 해당 연구는 먼저 다중 공선성 문제를 변수 제거 및 결합으로 해결하였다. 이후 의미있는 독립변수들을 뽑아내는 전진선택법(Forward Selection), 후진소거법(Backward Elimination), 단계적선택법(Stepwise Selection), L1 Regularization, 주성분분석(PCA) 총 5개의 변수 선택 알고리즘을 사용했다. 또한 심층신경망(DNN), XGBoost, CatBoost, Linear Regression 총 4개의 머신러닝 및 딥러닝 알고리즘을 이용해 하이퍼파라미터 최적화 후 모델을 학습시키고 모형간 예측력을 비교하였다. 추가 실험에서는 DNN의 node와 layer 수를 바꿔가면서 실험을 진행하여 가장 적절한 node와 layer 수를 찾고자 하였다. 결론적으로 가장 성능이 우수한 모델로 2021년의 아파트 매매실거래가격을 예측한 후 실제 2021년 데이터와 비교한 결과 훌륭한 성과를 보였다. 이를 통해 머신러닝과 딥러닝은 다양한 경제 상황 속에서 투자자들이 주택을 구매할 때 올바른 판단을 할 수 있도록 도움을 줄 수 있을 것이라 확신한다.

• Genomics & Informatics
• /
• 제14권4호
• /
• pp.149-159
• /
• 2016

With the success of the genome-wide association studies (GWASs), many candidate loci for complex human diseases have been reported in the GWAS catalog. Recently, many disease prediction models based on penalized regression or statistical learning methods were proposed using candidate causal variants from significant single-nucleotide polymorphisms of GWASs. However, there have been only a few systematic studies comparing existing methods. In this study, we first constructed risk prediction models, such as stepwise linear regression (SLR), least absolute shrinkage and selection operator (LASSO), and Elastic-Net (EN), using a GWAS chip and GWAS catalog. We then compared the prediction accuracy by calculating the mean square error (MSE) value on data from the Korea Association Resource (KARE) with body mass index. Our results show that SLR provides a smaller MSE value than the other methods, while the numbers of selected variables in each model were similar.

• 한국약용작물학회지
• /
• 제10권4호
• /
• pp.263-268
• /
• 2002

Acifluorfen-tolerant callus lines of Solanum ptycanthum were isolated by stepwise selection. Growth of the unselected line was completely inhibited at 0.5 uM. while some selected lines grew at 8 uM acifluorfen. Twenty-two of twenty-five acifluorfen-tolerant callus lines regenerated shoots. Many of the regenerated somaclones were variants, differing in leaf shape, leaf color, number of flower parts, flower color, and fertility. The acifluorfen tolerant S. ptycanthum callus lines differed.

• Communications for Statistical Applications and Methods
• /
• 제28권5호
• /
• pp.537-551
• /
• 2021

We conducted a study on a regression spline estimator with a few pre-specified auxiliary variables. For the implementation of the proposed estimators, we adapted a coordinate descent algorithm. This was implemented by considering a structure of the sum of the residuals squared objective function determined by the B-spline and the auxiliary coefficients. We also considered an efficient stepwise knot selection algorithm based on the Bayesian information criterion. This was to adaptively select smoothly functioning estimator data. Numerical studies using both simulated and real data sets were conducted to illustrate the proposed method's performance. An R software package psav is available.

• 한국작물학회지
• /
• 제42권5호
• /
• pp.539-547
• /
• 1997

This study was carried out to identify how soybean seed protein concentration is influenced by climatic factors. Twelve lines selected for seed protein concentration were studied in 13 environments of North Carolina. Sensitivity of seed protein concentration, total seed protein, and seed yield to climatic variables was investigated using a linear regression model. Best response models were determined using two stepwise selection methods, Maximum R-square and Stepwise Selection. There were wide climatic effects in seed protein concentration, total protein and seed yield. The highest protein concentration environment was characterized by the most high temperature days(HTD) and the smallest variance of average daily temperature range (VADTRg), while the lowest protein concentration environment was distinguished by the fewest HTD and the largest VADTRg. For protein concentration, all lines responded positively to average maximum daily temperature(MxDT), HTD, and average daily temperature range(ADTRg) and negatively to ADRa, while they responded positively or negatively to average daily temperature(ADT), variance of average minimum daily temperature (VMnDT), and VADTRg, indicating that genotypes may greatly differ in degrees of sensitivity to each climatic variable. Eleven lines seemed to have best response models with 2 or 3 variables. Exceptionally, NC106 did not show a significant sensitivity to any climatic variable and thus did not have a best response model. This indicates that it may be considered phenotypically more stable. For total seed protein and seed yield, all the lines responded negatively to both ADTRg and VADRa, suggesting that synthesis of seed components may increase with less daily temperature range and less variation in daily rainfall.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 1999년도 추계학술대회-지능형 정보기술과 미래조직 Information Technology and Future Organization
• /
• pp.365-373
• /
• 1999

Recently, numerous studies have demonstrated that artificial intelligence such as neural networks can be an alternative methodology for classification problems to which traditional statistical methods have long been applied. In building neural network model, the selection of independent and dependent variables should be approached with great care and should be treated as a model construction process. Irrespective of the efficiency of a learning procedure in terms of convergence, generalization and stability, the ultimate performance of the estimator will depend on the relevance of the selected input variables and the quality of the data used. Approaches developed in statistical methods such as correlation analysis and stepwise selection method are often very useful. These methods, however, may not be the optimal ones for the development of neural network models. In this paper, we propose a genetic algorithms approach to find an optimal or near optimal input variables for neural network modeling. The proposed approach is demonstrated by applications to bankruptcy prediction modeling. Our experimental results show that this approach increases overall classification accuracy rate significantly.

• 한국수자원학회논문집
• /
• 제55권4호
• /
• pp.257-266
• /
• 2022

국지성 호우 및 설계빈도 이상 강우의 증가로 침수피해가 매년 증가하고 있으며 이에 따라 홍수 조절 및 방어를 위한 수공구조물의 중요성이 증가하고 있다. 수공구조물은 목적과 성능에 따른 설계가 이루어지고 있고 홍수량이 중요한 산정 요소이나 국내에서는 관측자료의 신뢰성 부족 및 데이터의 부족으로 인하여 수공구조물 설계를 위한 수문해석 입력자료로 사용되는 설계강우량은 정확한 확률강우량의 산정과 시간분포가 중요한 요소로 작용한다. 실무에서는 Huff의 4분위 방법의 누가우량백분율을 이용하여 설계강우량의 시간분포 회귀식을 산정하고 있으며 분위별 곡선에 대한 회귀식은 전반적으로 정확도가 높게 나타나는 6차 다항회귀식을 일률적으로 사용하고 있다. 본 연구에서는 실무에서 일반적으로 설계강우량의 시간분포를 위해 사용하고 있는 Huff의 4분위 방법의 누가우량백분율을 이용하여 통계 모델링에서 간결함의 원리에 따라 변수선택법을 이용하여 시간분포 회귀식을 유도하였으며, 유의성 검정을 통한 시간분포 회귀식의 검증을 실시하였다. 변수선택법과 유의성 검정을 통한 시간분포 회귀식 산정 결과 전진선택법과 후방제거법의 장점을 모두 가지고 있는 단계선택법을 이용하여 시간분포 회귀식을 유도하는 것이 가장 적합한 것으로 분석되었다.