• The Pure and Applied Mathematics
• /
• v.13 no.3 s.33
• /
• pp.227-236
• /
• 2006

The purpose of this paper is to establish a random fixed point theorem for nonconvex valued random multivalued operators, which generalize known results in the literature. We also derive a random coincidence fixed point theorem in the noncompart setting.

• Journal of Environmental Science International
• /
• v.26 no.4
• /
• pp.521-527
• /
• 2017

Solar radiation forecasts are important for predicting the amount of ice on road and the potential solar energy. In an attempt to improve solar radiation predictability in Jeju, we conducted machine learning with various data mining techniques such as tree models, conditional inference tree, random forest, support vector machines and logistic regression. To validate machine learning models, the results from the simulation was compared with the solar radiation data observed over Jeju observation site. According to the model assesment, it can be seen that the solar radiation prediction using random forest is the most effective method. The error rate proposed by random forest data mining is 17%.

• Honam Mathematical Journal
• /
• v.27 no.3
• /
• pp.505-513
• /
• 2005

The aim of this paper is to establish a functional central limit theorem for multivariate moving average process generated by negatively associated random vectors under the finite second moments.

• The Pure and Applied Mathematics
• /
• v.24 no.4
• /
• pp.211-226
• /
• 2017

In this paper, machine learning models employed in various fields are discussed and applied to KOSPI200 stock index return forecasting. The results of hyperparameter analysis of the machine learning models are also reported and practical methods for each model are presented. As a result of the analysis, Support Vector Machine and Artificial Neural Network showed a better performance than k-Nearest Neighbor and Random Forest.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.2
• /
• pp.363-369
• /
• 2010

In this paper we propose a mixed-effects least squares support vector regression (LS-SVR) for longitudinal data. We add a random-effect term in the optimization function of LS-SVR to take random effects into LS-SVR for analyzing longitudinal data. We also present the model selection method that employs generalized cross validation function for choosing the hyper-parameters which affect the performance of the mixed-effects LS-SVR. A simulated example is provided to indicate the usefulness of mixed-effect method for analyzing longitudinal data.

• Communications for Statistical Applications and Methods
• /
• v.15 no.3
• /
• pp.343-351
• /
• 2008

By supervised learning with p predictors, we frequently obtain a prediction function of the form $y\;=\;f(x_1,...,x_p)$. When $p\;{\geq}\;3$, it is not easy to understand the inner structure of f, except for the case the function is formulated as additive. In this study, we propose to use p simple graphs for visual understanding of complex prediction functions produced by several supervised learning engines such as LOESS, neural networks, support vector machines and random forests.

• Genomics & Informatics
• /
• v.5 no.1
• /
• pp.1-5
• /
• 2007

Embryonic stem cells can be differentiated into various types of cells, requiring a tight regulation of transcription. Biomarkers related to each lineage of cells are used to guide the differentiation into neural or any other fates. In previous experiments, we reported the guided differentiation (GD)-specific genes by comparing profiles of random differentiation (RD). Interestingly 68% of differentially expressed genes in GD overlap with that of RD, which makes it difficult for us to separate the lineages by examining several markers. In this paper, we design a prediction model to identify the differentiation into neural fates from any other lineage. From the profiles of 11,376 genes, 203 differentially expressed genes between neural and random differentiation were selected by random variance T-test with 95% confidence and 5% false discovery rate. Based on support vector machine algorithm, we could select 79 marker genes from the 203 informative genes to construct the optimal prediction model. Here we propose a prediction model for the prediction of neural fates from random differentiation which is constructed with a perfect accuracy.

• Journal of the Korean Statistical Society
• /
• v.32 no.1
• /
• pp.11-20
• /
• 2003

Let｛Ｘt｝be an m-dimensional linear process of the form (equation omitted), where｛Zt｝is a sequence of stationary m-dimensional weakly associated random vectors with EZt = O and E∥Zt∥$^2$＜$\infty$. We Prove central limit theorems for multivariate linear processes generated by weakly associated random vectors. Our results also imply a functional central limit theorem.

• Proceedings of the KIPE Conference
• /
• 2001.10a
• /
• pp.160-168
• /
• 2001

The RPWM (Random· Pulse Width Modulation) is a switching technique to spread the voltage and current harmonics over a wide frequency area. By using randomly changing switching frequency of the inverter, the power spectrum of the electromagnetic acoustic noise can be spread to the wide-band area. The wide­band noise is much more comfortable and less annoying than the narrow-band one. So, the RPWM has been attracting interest as an excellent method for the reduction of acoustic noise on the inverter drive system. In this paper a new RPPWM (Random Position Space Vector PWM) is proposed and implemented. Each of three phase pulses is located randomly in each switching interval. Along with the randomization of PWM pulses, the space vector modulation is also executed in the C167 micro-controller. The experimental results show that the voltage and current harmonics are spread to a wide band area and that the audible switching noise is reduced by the proposed RPPWM method.

• The Korean Journal of Applied Statistics
• /
• v.31 no.2
• /
• pp.265-285
• /
• 2018

In this study, we applied a deep neural network model to predict four grades of fine dust $PM_{10}$, 'Good, Moderate, Bad, Very Bad' and two grades, 'Good or Moderate and Bad or Very Bad'. The deep neural network model and existing classification techniques (such as neural network model, multinomial logistic regression model, support vector machine, and random forest) were applied to fine dust daily data observed from 2010 to 2015 in six major metropolitan areas of Korea. Data analysis shows that the deep neural network model outperforms others in the sense of accuracy.