• Journal of the Korea Society of Computer and Information
• /
• v.18 no.4
• /
• pp.19-26
• /
• 2013

Pseudo sample neural network (PSNN) is a variant of traditional neural network using pseudo samples to mitigate the local-optima-convergence problem when the size of training samples is small. PSNN can take advantage of the smoothed solution space through the use of pseudo samples. PSNN has a focus on the quantity problem in training, whereas, methods stressing the quality of training samples is presented in this paper to improve further the performance of PSNN. It is evident that typical samples and highly correlated features help in training. In this paper, therefore, kernel density estimation is used to select typical samples and correlation factor is introduced to select features, which can improve the performance of PSNN. Debris flow data set is used to demonstrate the usefulness of the proposed methods.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.11
• /
• pp.11-18
• /
• 2012

Debris flow deposition model is a model to predict affected areas by debris flow and random walk model (RWM) was used to build the model. Although the model was proved to be effective in the prediction of affected areas, the model has several free parameters decided experimentally. There are several well-known methods to estimate parameters, however, they cannot be applied directly to the debris flow problem due to the small size of training data. In this paper, a modified neural network, called pseudo sample neural network (PSNN), was proposed to overcome the sample size problem. In the training phase, PSNN uses pseudo samples, which are generated using the existing samples. The pseudo samples smooth the solution space and reduce the probability of falling into a local optimum. As a result, PSNN can estimate parameter more robustly than traditional neural networks do. All of these can be proved through the experiments using artificial and real data sets.

• Communications for Statistical Applications and Methods
• /
• v.10 no.2
• /
• pp.479-496
• /
• 2003

We investigate logit confidence intervals for the odds ratio based on the delta method. These intervals are constructed using pseudo-Bayes estimators. The Gart method and Agresti method smooth the observed counts toward the model of equiprobability and independence, respectively. We obtain better coverage probability by smoothing the observed counts toward the pseudo-Bayes estimators in 2$\times$2 table. We also improve legit confidence intervals in 2$\times$2$\times$K tables by generalizing these ideas. Utilizing pseudo-Bayes estimators, we obtain better coverage probability by smoothing the observed counts toward the conditional independence model, no three-factor interaction model and saturated model in 2$\times$2$\times$K tables.

• Food Science and Biotechnology
• /
• v.14 no.6
• /
• pp.836-840
• /
• 2005

Extraction characteristics of Bonus species of Brassica oleracea var. capita and functional properties of corresponding extract were monitored by response surface methodology (RSM). Maximum extraction yield of 44.07% was obtained at ratio of solvent to sample of 27.94 mL/g, ethanol concentration of 24.35%, and extraction temperature of $55.21^{\circ}C$. At ratio of solvent to sample, ethanol concentration, and extraction temperature of 21.11 mL/g, 58.53%, and $68.83^{\circ}C$, respectively, maximum electron-donating ability was 48.44%. Maximum inhibitory effect on tyrosinase was 68.94% at ratio of solvent to sample, ethanol concentration, and extraction temperature of 24.08 mL/g, 10.49%, and $78.71^{\circ}C$, respectively. Superoxide dismutase (SOD) showed maximum pseudo-activity of 24.78% at ratio of solvent to sample of 22.66 mL/g, ethanol concentration of 45.69%, and extraction temperature of $93.81^{\circ}C$. Based on superimposition of four-dimensional RSM with respect to extraction yield, electron-donating ability, and pseudo-activity of SOD, optimum ranges of extraction conditions were ratio of solvent to sample of 20-30 mL/g, ethanol concentration of 35-65%, and extraction temperature of $50-80^{\circ}C$.

• The Korean Journal of Food And Nutrition
• /
• v.6 no.2
• /
• pp.109-114
• /
• 1993

This study was divised to observe an Inhibitory effect toward a lipolytic action of toxohormone-L from large root and small root Nepal pseudo ginseng (NPG ; Nepal products) components by water extract and ethanol precipitate in vitro. Toxohormone-L is known to be a lipolytic factor that was partially purified from the ascites fluid of Sarcoma 180-bearing mice and of patients with hepatoma. The inhibitory effect that inhibited the lipolytic action of toxohormone-L by ethanol Precipitate component of large root NPG (mean 55.5%) was higher (mean 1.37 times) than that of water extract component in final reaction concentration of 500 and 1, 000ug/ml, on the other side inhibitory effect of water extract component in small root NPG (mean 55.5%) was higer (mean 1.14 times) than that of ethanol precipitate component. In a way inhibitory effect of precipitate component In large root NPG(47.6%), when final reaction concentration of sample were 1, 000ug/ml, was about 40% lower than that of Korean red ginseng.

• The Korean Journal of Financial Management
• /
• v.24 no.3
• /
• pp.91-131
• /
• 2007

In the paper, the effects of sidecar on index arbitrage trading and non-index arbitrage trading in the Korean stock market are examined. The analyses of return, volatility, and liquidity dynamics illustrate that there are no distinct differences for index arbitrage group and non-index arbitrage group surrounding the sidecar events. For further analysis, we construct pseudo-sidecar sample and analyse the effects of the actual sidecar and pseudo-sidecar on arbitrage sample and non-index arbitrage sample. The result of analysis using pseudo-sidecar shows that the differences between index arbitrage group and non-index arbitrage group are larger in pseudo-sidecar sample than in actual sidecar sample. This means that former results can be explained by temporary order clustering in one side before and after the event. Sidecar has little effect on non-index arbitrage group, however, it has relatively large effect on arbitrage group. These results imply that it needs to redesign the sidecar system of the Korean stock market which applies for all program trading including arbitrage and non-index arbitrage trading.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2013.01a
• /
• pp.197-199
• /
• 2013

신경망의 학습은 학습 샘플의 품질뿐만이 아니라 입력으로 사용되는 특징에도 영향을 받으므로 신경망의 출력을 결정하는데 있어 연관성이 높은 특징을 입력으로 사용함으로써 학습된 신경망의 전체적인 성능을 높일 수 있다. 이 논문에서는 신경망의 입력으로 사용되는 특징과 출력의 연관성 파악하고 연관성이 낮은 특징을 학습 과정에서 배제함으로써 신경망의 전체적인 성능을 높일 수 있는 방법을 제시하였다. 토석류 데이터를 위한 의사 샘플 신경망에 제안한 방법을 적용한 경우 연관성이 낮은 특징 하나를 제외함으로써 약 6%의 오류 감소 효과를 얻을 수 있었다.

• The Korean Journal of Food And Nutrition
• /
• v.4 no.1
• /
• pp.75-80
• /
• 1991

This study was divised to observe an inhibitory toward a lipolytic action of toxohormone-L from large root and small root Nepal pseudo ginseng(NPG ; Nepal products) components by water extract and ethanol precipitate in vitro. Toxohormone-L Is known to be a lipolytic factor that was partially purified from the ascites fluid of sarcoma 180-hearing mice and of patients with hepatoma. The inhibitory effect that inhibited the lipolytic action of toxohormone-L by ethanol precipitate component of large root NPG(mean 46.8%) was higher (mean 1.8 times) than that of water extract component in final reaction concentration ,5001g1m1, on the other side inhibitory effect of water extract component in small root NPG(mean 43.9%) was higher(mean 1. 2 times) than that of ethano1 precipitate component, respectively. In a way inhibitory effect of ethanol precipitate component in large root NPG(47.6%), when final reaction concentration of sample were 1,000 U g/ml, was about 4095 lower than that of Korean red ginseng, respectively.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.6
• /
• pp.1557-1564
• /
• 2017

We take genomic sequences of high-dimensional low sample size (HDLSS) without ordering of response categories into account. When constructing an appropriate test statistics in this model, the classical multivariate analysis of variance (MANOVA) approach might not be useful owing to very large number of parameters and very small sample size. For these reasons, we present a pseudo marginal model based upon the Gini-Simpson index estimated via Bayesian approach. In view of small sample size, we consider the permutation distribution by every possible n! (equally likely) permutation of the joined sample observations across G groups of (sizes $n_1,{\ldots}n_G$). We simulate data and apply false discovery rate (FDR) and positive false discovery rate (pFDR) with associated proposed test statistics to the data. And we also analyze real SARS data and compute FDR and pFDR. FDR and pFDR procedure along with the associated test statistics for each gene control the FDR and pFDR respectively at any level ${\alpha}$ for the set of p-values by using the exact conditional permutation theory.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.40 no.1
• /
• pp.15-23
• /
• 2003

This paper proposes a fast direction of arrival (DOA) estimation algorithm that can rapidly estimate incidence angles of incoming signals using a pseudo covariance matrix. The conventional subspace DOA estimation methods such as MUSIC (multiple signal classification) algorithms need many sample signals to acquire covariance matrix of input signals. Thus, it is difficult to estimate the DOAs of signals because they cannot perform DOA estimation during receiving sample signals. Also if the D0As of signals are changing rapidly, conventional algorithms cannot estimate incidence angles of signals exactly. The proposed algorithm obtains bearing response and directional spectrum after acquiring pseudo covariance matrix of each snapshot. The incidence angles can be exactly estimated by using the bearing response and directional spectrum. The proposed DOA estimation algorithm uses only concurrent snapshot so as to obtain covariance matrix. Compared to conventional DOA estimation methods. The proposed algorithm has an advantage that can estimate DOA of signal rapidly.