• Communications for Statistical Applications and Methods
• /
• v.16 no.1
• /
• pp.185-193
• /
• 2009

Machine learning methods such as support vector machines and random forests yield nonparametric prediction functions of the form y = $f(x_1,{\ldots},x_p)$. As a sequel to the previous article (Huh and Lee, 2008) for visualizing nonparametric functions, I propose more sensible graphs for visualizing y = $f(x_1,{\ldots},x_p)$ herein which has two clear advantages over the previous simple graphs. New graphs will show a small number of prototype curves of $f(x_1,{\ldots},x_{j-1},x_j,x_{j+1}{\ldots},x_p)$, revealing statistically plausible portion over the interval of $x_j$ which changes with ($x_1,{\ldots},x_{j-1},x_{j+1},{\ldots},x_p$). To complement the visual display, matching importance measures for each of p predictor variables are produced. The proposed graphs and importance measures are validated in simulated settings and demonstrated for an environmental study.

• Journal of the Korean Geophysical Society
• /
• v.7 no.4
• /
• pp.277-282
• /
• 2004

Creep deformation of concrete is often responsible for excessive deflection at loads which can compromise the performance of elements within structures. Hence, the prediction of the magnitude and rate of creep strain is an important requirement of the design process and management of structures. Although laboratory tests may be undertaken to determine the deformation properties of concrete, these are time-consuming, often expensive and generally not a practical option. Therefore, relatively simple empirically based national design code models are relied upon to predict the magnitude of creep strain.This paper reviews the accuracy of creep predictions yielded by eight commonly used international "code type" models, all of which do not consider the same material parameters and yield a range of predicted strains, when compared with actual strains measured on a range of concretes in seventeen different investigations. The models assessed are the: SABS 0100 (1992), BS 8110 (1985), ACI 209 (1992), AS 3600 (1998), CEB-FIP (1970, 1978 and 1990) and the RILEM Model B3 (1995). The RILEM Model B3 (1995) and CEB-FIP (1978) were found to be the most and least accurate, respectively.

• New & Renewable Energy
• /
• v.7 no.4
• /
• pp.3-9
• /
• 2011

Optimization of initial total sugar concentration of sweet sorghum juice, aeration time and aeration rate on ethanol production was performed by response surface methodology (RSM). The optimum conditions for ethanol production from concentrated sweet sorghum juice were determined as follows: initial total sugar concentration, 21.2 Brix; aeration time, 7.66h; aeration rate, 1.22 vvm. At the optimum conditions, the maximum ethanol yield was predicted to be 91.65% by model prediction. Similarly, 92.98% of ethanol yield was obtained by verification experiment using optimum conditions after 48 h of fermentation. This result was in agreement with the model prediction.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.15 no.2
• /
• pp.59-66
• /
• 2013

Among many factors, especially meteorological conditions can impact agricultural productivities. This study was conducted to analyze the relationships between crop yield and meteorological factors. We collected meteorological data (i.e., temperature and precipitation) from the Automated Weather System (AWS) of Korea Meteorological Administration (KMA) and the yield data of Chinese cabbage and Radish from local Nonghyup (NCAF:National Agricultural Cooperative Federation) and Farmers' Corporate Association. The agricultural data were classified into two groups. These groups are comprised of the farmers who produced a crop under 30 kg per $3.3m^2$ and over 30k g per $3.3m^2$ respectively. The daily meteorological data were calculated from the average value for ten days. Based on the regression analysis, we concluded that the yield of Chinese cabbage (Haenam) was related to average temperature, minimum temperature, precipitation, and number of days with precipitation, whereas that of Radish (Jeju) was related to average temperature, maximum temperature, and minimum temperature. The result suggests that these meteorological data can be used more effectively for the prediction of crop yield.

• Korean Journal of Weed Science
• /
• v.30 no.4
• /
• pp.390-395
• /
• 2010

This study was conducted to predict reduction of soybean yield as affected by different densities of Cuscuta pentagona. All data were fitted to Cousens' rectangular hyperbola model to estimate parameters for predicting soybean yield loss. The yield of soybean in the various densities (1 to 48 plants $m^{-2}$) of C. pentagona reduced by 80.3 to 99.7%, respectively. Among yield components, number of pods was the most significantly influenced by weed interferences. The prediction model for soybean yield as affected by weed competition was as follows: Y= 274.6783/(1+4.3522X), $r^2$=0.999 in C. pentagona. Economic threshold levels calculated using cousens' equation was 0.004 plants $m^{-2}$ in C. pentagona.

• Journal of Korean Society of Forest Science
• /
• v.99 no.4
• /
• pp.479-486
• /
• 2010

We analyzed the relationships between the daily yield and quality of pine-mushroom to predict the annual production pattern and unit price of the mushroom with the records of pine-mushroom trade at Yeongdeok forestry cooperative's market for nine years (2000~2008). Although there were some exceptions due to extreme drought or extraordinary temperature, the production ratio of high quality (first and second grade) was more than 50% in early stage and decreased, while that of low quality (pileus opened and defected ones) showed increasing pattern after the production reached in peak. The ratio of high quality and that of low quality were reversed 1~9 days before the mushroom production reached the acme of daily yield, which allowed us to predict that the mushroom production would be decreased when the ratio of low quality overcomes that of high quality. The ratio of high quality preceded about 3~4 days prior to that of daily yield, and the mushroom yield showed significant correlations with the ratio of high quality mushroom prior to 3~4 days of the day with the coefficient larger than 0.5 (r=0.51 for 3 days and r=0.54 for 4 days). Thus, we concluded that the analysis of grade distribution of pine-mushroom at the market may provide a significant clue to predict production pattern of the mushroom. In addition, the price of high quality pine-mushroom showed clear negative correlations with the yield. Thus, the analysis may take a good role for the trading of pine-mushroom with providing information for predicting the price of pine-mushroom.

• Weed & Turfgrass Science
• /
• v.1 no.4
• /
• pp.31-37
• /
• 2012

This study was conducted to make the rice yield prediction model system as affected by densities of Aneilema keisak and Aeschynomene indica and to determine their economic threshold levels in flooded direct-seeded rice. When the density of A. keisak was 8 plants per $m^2$, the yield of rice reduced to 8% and as the density increased up to 96 plants per $m^2$, the reduced rate of rice yield reached to 45% and in A. indica, the reduced rate of rice yield were 20 and 77%, respectively. The rice yield loss models of A. keisak and A. indica were predicted as Y=553.2 kg (1+0.00913X), $R^2=0.912^{**}$ and Y=567.9 kg/(1+0.04434X), $R^2=0.961^{**}$, respectively. Economic threshold levels calculated using cousens' equation were 3.0 plants per $m^2$ in A. keisak and 0.6 plants per $m^2$ in A. indica.

• Asian-Australasian Journal of Animal Sciences
• /
• v.33 no.12
• /
• pp.1912-1921
• /
• 2020

Objective: This study assessed genomic prediction accuracies based on different selection methods, evaluation procedures, training population (TP) sizes, heritability (h²) levels, marker densities and pedigree error (PE) rates in a simulated Korean beef cattle population. Methods: A simulation was performed using two different selection methods, phenotypic and estimated breeding value (EBV), with an h² of 0.1, 0.3, or 0.5 and marker densities of 10, 50, or 777K. A total of 275 males and 2,475 females were randomly selected from the last generation to simulate ten recent generations. The simulation of the PE dataset was modified using only the EBV method of selection with a marker density of 50K and a heritability of 0.3. The proportions of errors substituted were 10%, 20%, 30%, and 40%, respectively. Genetic evaluations were performed using genomic best linear unbiased prediction (GBLUP) and single-step GBLUP (ssGBLUP) with different weighted values. The accuracies of the predictions were determined. Results: Compared with phenotypic selection, the results revealed that the prediction accuracies obtained using GBLUP and ssGBLUP increased across heritability levels and TP sizes during EBV selection. However, an increase in the marker density did not yield higher accuracy in either method except when the h² was 0.3 under the EBV selection method. Based on EBV selection with a heritability of 0.1 and a marker density of 10K, GBLUP and ssGBLUP_0.95 prediction accuracy was higher than that obtained by phenotypic selection. The prediction accuracies from ssGBLUP_0.95 outperformed those from the GBLUP method across all scenarios. When errors were introduced into the pedigree dataset, the prediction accuracies were only minimally influenced across all scenarios. Conclusion: Our study suggests that the use of ssGBLUP_0.95, EBV selection, and low marker density could help improve genetic gains in beef cattle.

• Journal of Welding and Joining
• /
• v.26 no.3
• /
• pp.61-66
• /
• 2008

As ships become to be larger than ever, the thicker plate and the higher tensile steel plate are used in naval shipyard. Though special chemical composition is needed for high-tensile steels, recent high-tensile steels are made by the TMCP(Thermo-Mechanical control process) skill. The increase of yield stress and tensile stress of TMCP steels is induced from bainite phase which is transformed from austenite, but that increased yield stress can be vanished by another additional thermal cycle like welding and heating. As thermal deformations are deeply related by yield stress of material, the study for prediction of plate deformation by heating should reflect principle of TMCP steels. This study developed an algorithm which can calculate inherent strain. In this algorithm, not only the mechanical principles of thermal deformations, but also the predicting of the portion of initial bainite is considered when calculating inherent strain. The simulations of plate deformation by these values showed good agreements with experimental results of normalizing steels and TMCP steels in welding and heating. Finally we made an inherent strain database of steels used in Class rule.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.53 no.3
• /
• pp.75-82
• /
• 2011

This study refers to the derivation of simple crop yield prediction equation by using KOMPSAT-2 derived vegetation index. For a 1.25 ha small farm area located in the middle part of South Korea, the KOMPSAT-2 panchromatic and multi-spectral images of 31th August 2008, 17th November 2008, and 10th September 2009 were used. The field spectral reflectance during growing period for the 6 crops (rice, potato, corn, red pepper, garlic, and bean) were measured using ground spectroradiometer and the yield was investigated. Among the 6 vegetation indices (VI), the NDVI and ARVI between measured and image derived showed high relationship with the coefficient of determination of 0.85 and 0.95 respectively. Using the 3 years field data, the NDVI and ARVI regression curves were derived, and the yields were tried to compare with the maximum VIs value.