• The Korean Journal of Applied Statistics
• /
• v.18 no.3
• /
• pp.661-671
• /
• 2005

The aim of this study is to propose a clustering method, called tree-structured clustering, by recursively partitioning continuous multivariate dat a based on overall $R^2$ criterion with a practical node-splitting decision rule. The clustering method produces easily interpretable clustering rules of tree types with the variable selection function. In numerical examples (Fisher's iris data and a Telecom case), we note several differences between tree-structured clustering and K-means clustering.

• Proceedings of the Korean Statistical Society Conference
• /
• 2005.05a
• /
• pp.49-51
• /
• 2005

본 연구는 반복분할(recursive partitioning)에 의한 군집화 방법을 제안하고 활용 예를 제시한다. 이 방법은 나무 형태의 해석하기 쉬운 단순한 규칙을 제공하면서 동시에 변수선택기능을 제공한다.

• Journal of Biosystems Engineering
• /
• v.23 no.6
• /
• pp.571-582
• /
• 1998

The PLS(Partial Least Square) and ANN(Artificial Neural Network) were introduced to develop the soluble solids content prediction model of apples which is followed by making a subsequent selection of photosensor. For the optimal PLS model, number of factors needed for spectrum analysis were increased until the convergence of prediction residual error sum of squares. Analysis has shown that even part of the overall wavelength with no pretreatment may turn out better performing. The best PLS model was found in the 800 to 1,100nm wavelength region without pretreatment of second derivation, having $R^2$=0.9236, bias= -0.0198bx, SEP=0.2527bx for unknown samples. On the other hand, for the ANN model the second derivation led to higher performance. On partial range of 800 to 1,100nm wavelengh region, prediction model with second derivation for unknown samples reached $R^2$=0.9177, SEP=0.2903bx in contrast to $R^2$=0.7507, SEP =0.4622bx without pretreatment.

• The Korean Journal of Financial Management
• /
• v.6 no.2
• /
• pp.33-69
• /
• 1989

The Purpose of this study is to test empirically the determinants of capital structure of the Korea Listed Firms. In order to accomplish the purpose of this study, both literature survey and empirical test have been made. For the empirical test, agency and asymmetric information factors as well as traditional ones have been throughly reviewed. Traditional factors tested in this study include firm-size, collateral value of the assets, business risk, tax, non-debt tax shields, and industry effects. Agency and asymmetric information factors include growth opportunities, the percentage of outstanding equity held by inside stockholders, and the number of inside stockholders. From the results of the cross-sectional regression analysis, the adjusted R-square is 1931%, and the overall F-value indicates significance. For the analysis period, the signs of the variables except business risk are as predicted. Firm-size, collateral value of the assets, and business risk significant at the.01-.05 level. In order to determine the influence of industry factors on the financial leverage, a total of 8 dummy variables are added to the regression model. The adjusted R-square inclosed by 4.2% for the first analysis period(1983-1985) and 6% for the second analysis period(1986-1987). This suggests that industry factors are significant in explaining the variations in financial leverage across firms. In order to pursue the influence of agency and asymmetric information factors on the financial leverage, again the cross-sectional regression analysis is done for the middle size firms gruop(n=40). The adjusted R-square increased by 9.8% for the first analysis period(1983-1985) and 6.1% for the total analysis period(1983-1987), and all the signs was as predicted. But both the variables except the number of inside stockholders was not significant.

• ETRI Journal
• /
• v.34 no.4
• /
• pp.511-517
• /
• 2012

The best linear unbiased estimator (BLUE) is most suitable for practical application and can be determined with knowledge of only the first and second moments of the probability density function. Although the BLUE is an existing algorithm, it is still largely unexplored and has not yet been applied to channel estimation in amplify and forward (AF)-based wireless relay networks (WRNs). In this paper, a BLUE-based algorithm is proposed to estimate the overall channel impulse response between the source and destination of AF strategy-based WRNs. Theoretical mean square error (MSE) performance for the BLUE is derived to show the accuracy of the proposed channel estimation algorithm. In addition, the Cram$\acute{e}$r-Rao lower bound (CRLB) is derived to validate the MSE performance. The proposed BLUE channel estimation algorithm approaches the CRLB as the length of the training sequence and number of relays increases. Further, the BLUE performs better than the linear minimum MSE estimator due to the minimum variance characteristic exhibited by the BLUE, which happens to be a function of signal-to-noise ratio.

• Journal of Korean Academy of Nursing Administration
• /
• v.16 no.1
• /
• pp.86-100
• /
• 2010

Purpose: This study was done to examine the relationship of job satisfaction and organizational commitment of perioperative nurses. Method: The subjects of this study were 500 perioperative nurses from 11 hospitals. The data were collected by self-reporting questionnaires from Sep. 19 to Sep. 27, 2009. Results: There was statistically significant relationship among the five variables. The analyses of covariance of these five variables revealed overall significant (p<.05). Stepwise linear multiple regression analyses were used to examine the influence of these five variables. Results indicated that the variables for verbal abuse (p<.01), workplace climate (p<.01), internal marketing (p<.001), and job transfer (p<.001) contributed significantly to the job satisfaction (adjusted R square=.426), while the verbal abuse (p<.01), internal marketing (p<.01), leadership style (p<.001) and workplace climate (p<.001) did to the organizational commitment (adjusted R square=.351). Canonical correlation analyses revealed that internal marketing and workplace climate contributed most significantly both to job satisfaction and organizational commitment. Conclusion: This study found that all these five nursing managerial factors were important influential on both job satisfaction and organizational commitment of perioperative nurses. Addressing these factors with further research will surely improve the commitment of these nurses and ultimately lead to better perioperative nursing care.

• Preventive Nutrition and Food Science
• /
• v.18 no.1
• /
• pp.67-75
• /
• 2013

Descriptive analysis and consumer acceptance tests were conducted with frozen (FCR), homemade (HCR), and aseptic-packaged (ACR) cooked rice products from two cultivars-IM and SD. FCR was prepared using a rapid freezing process, which may provide consumers with a quality similar to that of HCR. The intensity of the flavors of roasted, glutinous rice, rice cake, and rice starch and the textures of glutinousness, moistness, chunkiness, adhesiveness, and squishiness were all greater in the FCR as compared to the HCR and ACR (p<0.05) in IM and SD cultivars. The differences in sensory characteristics between the FCR and ACR were larger than the equivalent differences between the FCR and HCR. Overall consumer acceptance ratings for FCR in overall aspect, appearance, aroma, and texture were not significantly different compared to those for HCR (p>0.05); however, in most cases these factors showed significant differences when compared with ACR (p<0.05). From partial least square regression analysis, cooked rice was positively related to sweet, transparency, glossiness, roasted, glutinousness, chunkiness, moistness, glutinous rice, adhesiveness, rice shape, rice starch, and squishiness attributes but negatively related to raw rice, old rice, old rice aroma, a particle feeling, off-aroma, white color, scatteredness, slickness, size of cooked rice, and firmness attributes.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.43 no.1
• /
• pp.128-134
• /
• 2014

To develop a crab-like flavorant from snow crab cooker effluent (SCCE, $20^{\circ}Brix$), optimal reaction conditions were determined using response surface methodology (RSM) combined with reaction flavoring technology (RFT). Using five variables (proline, glycine, arginine, methionine, fructose), RSM based on a five-level central composite design was applied to evaluate sensory acceptance (odor, taste, and overall acceptance) as dependent variables. A model equation obtained from RSM showed 0.88 of R-square for odor, 0.90 for taste, and 0.95 for overall acceptance with 0.07 lack of fit in overall acceptance (P<0.05). Odor score (predicted value) was 7.21 in the saddle point. Optimal flavoring conditions for making a crab-like flavorant were as follows: addition of 0.29 g of proline, 0.63 g of glycine, 0.61 g of arginine, 0.02 g of methionine, and 1.07 g% (w/v) of fructose into SCCE with RFT (90 min at $130^{\circ}C$). Odor score obtained under optimal conditions was 7.56, which was higher than the predicted value.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.4
• /
• pp.487-494
• /
• 2023

In this study, performance comparison was performed on R and TensorFlow, which are free deep learning tools. In the experiment, six types of deep neural networks were built using each tool, and the neural networks were trained using the 10-year Korean temperature dataset. The number of nodes in the input layer of the constructed neural network was set to 10, the number of output layers was set to 5, and the hidden layer was set to 5, 10, and 20 to conduct experiments. The dataset includes 3600 temperature data collected from Gangnam-gu, Seoul from March 1, 2013 to March 29, 2023. For performance comparison, the future temperature was predicted for 5 days using the trained neural network, and the root mean square error (RMSE) value was measured using the predicted value and the actual value. Experiment results shows that when there was one hidden layer, the learning error of R was 0.04731176, and TensorFlow was measured at 0.06677193, and when there were two hidden layers, R was measured at 0.04782134 and TensorFlow was measured at 0.05799060. Overall, R was measured to have better performance. We tried to solve the difficulties in tool selection by providing quantitative performance information on the two tools to users who are new to machine learning.

• Proceedings of the KSME Conference
• /
• 2001.06e
• /
• pp.450-455
• /
• 2001

An analytical study is made of transient adjustment process of an initially stationary, stably-stratified fluid in a square. The boundary walls are highly-conducting. The overall Rayleigh number $R_a$ is large. Considerations are given to both opposing (${\delta}w/{\delta}T>0$) and cooperating (${\delta}w/{\delta}T<0$) configurations. The flow character in opposing configuration can be classified into (a) a forced-convection dominaut mode (${\delta}w/{\delta}T>1/\sqrt{2}$), (b) a buoyancy-convection dominant mode ($0<{\delta}w/{\delta}T<1/\sqrt{2}$), and (c) a static mode (${\delta}w/{\delta}T{\cong}1/\sqrt{2}$). Global evolutionary processes are depicted. and physical rationalizations are provided.