• The Korean Journal of Applied Statistics
• /
• v.34 no.2
• /
• pp.177-190
• /
• 2021

Random forests is a popular method that improves the instability and accuracy of decision trees by ensembles. In contrast to increasing the accuracy, the ease of interpretation is sacrificed; hence, to compensate for this, variable importance is provided. The variable importance indicates which variable plays a role more importantly in constructing the random forests. However, when a predictor is correlated with other predictors, the variable importance of the existing importance algorithm may be distorted. The downward bias of correlated predictors may reduce the importance of truly important predictors. We propose a new algorithm remedying the downward bias of correlated predictors. The performance of the proposed algorithm is demonstrated by the simulated data and illustrated by the real data.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.6
• /
• pp.754-758
• /
• 2003

The present paper establishes a strong law of large numbers for sums of tight fuzzy random variables as a generalization of SLLN for sums of independent and identically distributed fuzzy random variables.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.7
• /
• pp.852-856
• /
• 2004

The present paper establishes a necessary and sufficient condition for weak convergence for weighted sums of compactly uniformly integrable level-continuous fuzzy random variables as a generalization of weak laws of large numbers for sums of fuzzy random variables.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.3021-3023
• /
• 1999

엄격한 시간 제약성에 의해 특성화되는 실시간 시스템의 성능을 평가하기 위해서 퍼지-랜덤 변수가 포함된 이산 사건 모델을 제시한다. 실시간 시스템의 정확성은 출력의 논리적 결과 뿐 아니라 반응시간에도 의존하므로, 본 논문에서는 실시간 시스템의 성능을 유연하게 평가하기 위해서 퍼지-랜덤 변수에 의해 적절하게 변형된 상태 오토마타를 제시하고 그 오토마타를 적용한 수치 예제를 제시한다.

• Proceedings of the KIPE Conference
• /
• 2015.07a
• /
• pp.497-498
• /
• 2015

Random PWM 방식은 기본파의 배수 고조파에 밀집된 성분을 광대역화할 수 있기 때문에 특정 주파수의 노이즈 감소 및 소음 저감 등에 효과가 있는 인버터 스위칭 방식이다. Random PWM을 구현하는 방법은 랜덤 변수를 생성하고 이를 이용하여 스위칭 주파수를 가변하거나, 인가되는 벡터의 위치를 변화시키는 등의 다양한 방법이 있다. 본 연구에서는 Random PWM 구현 방식, 랜덤 변수의 변동 주기, 랜덤 변수가 적용되는 크기 등의 차이에 따라 주파수 광대역화에 미치는 효과를 분석하였다. 이 결과로부터 구현을 위해 요구되는 조건에 따른 Random PWM의 효과를 비교할 수 있으며, 투입 비용 및 시스템 제한 조건 대비 최적의 Random PWM 방식을 선정할 수 있는 기준으로 활용될 수 있다.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.10 no.4
• /
• pp.349-360
• /
• 2008

In this paper, the modeling procedure of random field with an elasto-plastic finite element algorithm and probability of failure on closely-spaced tunnel were investigated. Local average subdivision (LAS) method which can generate discrete random variables fast and accurately as well as change the resolution in certain region was used. And correlated value allocating and weighted average method were suggested to implement geometrical characteristics of tunnel. After the probability of failure on the test problem was thoroughly investigated using random finite element method, the results were compared with the deterministic strength reduction factor method and single random variable method. Of particular importance in this work, is the conclusion that the probability of failure determined by simplified probabilistic analysis, in which spatial variability is ignored by assuming perfect correlation, can be estimated from the safety factor determined by strength reduction factor method. Also, single random variable method can lead to unconservative estimates of the probability of failure.

• The Korean Journal of Applied Statistics
• /
• v.34 no.2
• /
• pp.153-166
• /
• 2021

A main goal of pharmacogenomics studies is to predict individual's drug responsiveness based on high dimensional genetic variables. Due to a large number of variables, feature selection is required in order to reduce the number of variables. The selected features are used to construct a predictive model using machine learning algorithms. In the present study, we applied several hybrid feature selection methods such as combinations of logistic regression, ReliefF, TurF, random forest, and LASSO to a next generation sequencing data set of 400 epilepsy patients. We then applied the selected features to machine learning methods including random forest, gradient boosting, and support vector machine as well as a stacking ensemble method. Our results showed that the stacking model with a hybrid feature selection of random forest and ReliefF performs better than with other combinations of approaches. Based on a 5-fold cross validation partition, the mean test accuracy value of the best model was 0.727 and the mean test AUC value of the best model was 0.761. It also appeared that the stacking models outperform than single machine learning predictive models when using the same selected features.

• The Journal of Natural Sciences
• /
• v.17 no.1
• /
• pp.51-54
• /
• 2006

A complete convergence theorem for randomly weighted sums of random variables is obtained. No conditions are imposed on the joint distributions of the random variables.

• The Korean Journal of Applied Statistics
• /
• v.28 no.6
• /
• pp.1121-1131
• /
• 2015

Logistic models with a random intercept are useful to analyze longitudinal binary data. Traditionally, the random intercept of the logistic model is assumed to be parametric (such as normal distribution) and is also assumed to be independent to variables. Such assumptions are very strong and restricted for application to real data. Recently, Garcia and Ma (2015) derived semiparametric efficient estimators for logistic model with a random intercept without these assumptions. Their estimator shows the consistency where we do not assume any parametric form for the random intercept. In addition, the method is computationally simple. In this paper, we apply this method to analyze toenail infection data. We compare the semiparametric estimator with maximum likelihood estimator, penalized quasi-likelihood estimator and hierarchical generalized linear estimator.

• Proceedings of the KIEE Conference
• /
• 1999.11c
• /
• pp.794-796
• /
• 1999

엄격한 시간 제약성에 의해 특성화되는 실시간 전력 시스템의 성능 및 신뢰도를 평가하기 위해서 퍼지-랜덤 변수가 포함된 이산 사건 모델 및 확장된 path-space 기법을 제시한다. 실시간 시스템의 정확성은 출력의 논리적 결과 뿐 아니라 반응시간에도 의존하므로, 본 논문에서는 실시간 전력 시스템의 성능 및 신뢰도를 유연하게 평가하기 위해서 퍼지-랜덤 변수에 의해 적절하게 변형된 상태 오토마타를 제시하고 몇가지 수치 예제를 제시함으로써 제안한 기법의 효용성을 검증한다.