• Journal of Food Hygiene and Safety
• /
• v.16 no.3
• /
• pp.194-199
• /
• 2001

This study was designed to compare the effectiveness and applicability of the HApS$^{TM}$ (Hazard Analysis process System; HUKO, Seoul, Korea) based on Petrifilm$^{TM}$ (3M, St. Paul, MN, USA) with the AOAC (the Association of Official Analytical Chemists) standard total aerobic count (TAC) method and coliform count (CC) method for meat products. The comparisons were carried out using 230 meat samples collected from various retailers: 80 pork samples, 80 chicken samples, and 70 beef samples. In the comparison of the correlation coefficient (r) between conventional method and HApS$^{TM}$ method by a linear regression analysis, the correlation coefficients in total microorganism were 0.97767, 0.90712, and 0.95594 in pork, beef, and chicken samples, respectively. The correlation coefficients in coliform count were 0.82062, 0.94833, and 0.96839 in pork, beef and chicken samples, respectively. All the independent t-test on measurement values between conventional method and HApS$^{TM}$ method represented no significant differences in the means between two methods at the 0.05 of significance level($\alpha$=0.05). Based on the high correlation between HApS$^{TM}$ and the AOAC standard methods in the TAC and CC, it might be compatible to employ the HApS$^{TM}$ method to measure the microbial contamination in livestock products. HApS$^{TM}$ method was simpler and less time-consuming in sample preparation and procedures faster than the conventional method. These results suggested that the HApS$^{TM}$ method could be substitute for the conventional methods in the analysis of microbial contamination measurement in meat products.n meat products.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.3 s.309
• /
• pp.25-33
• /
• 2006

This paper proposed device-independent color matching algorithm based on the 3D RGB lookup table (LUT) between mobile camera and mobile LCD (Liquid Crystal Display) to improve the color-fidelity. Proposed algorithm is composed of thee steps, which is device characterization, gamut mapping, 3D RGB-LUT design. First, the characterization of mobile LCD is executed using the sigmoidal function, different from conventional method such as GOG (Gain Offset Gamma) and S-curve modeling, based on the observation of electro-optical transfer function of mobile LCD. Next, mobile camera characterization is conducted by fitting the digital value of GretagColor chart captured under the daylight environment (D65) and tristimulus values (CIELAB) using the polynomial regression. However, the CIELAB values estimated by polynomial regression exceed the maximum boundary of the CIELAB color space. Therefore, these values are corrected by linear compression of the lightness and chroma. Finally, gamut mapping is used to overcome the gamut difference between mobile camera and moible LCD. To implement the real-time processing, 3D RGB-LUT is designed based on the 3D RGB-LUT and its performance is evaluated and compared with conventional method.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.3
• /
• pp.677-687
• /
• 2016

Among lots of sabermetric statistics for baseball batters' ability, the wins above replacement (WAR) is the most popular statistic in MLB. However, there exists a difficulty applying WAR to KBO, since KBO data do not have position adjustment, league adjustment and park factor which are essential in calculating WAR. In this paper, using five statistics for both KBO and MLB qualified batters, we propose hitting ability index (HAI), an alternative sabermetric indices to represent batters' ability. Comparing HAI with WAR of MLB batters, we evaluate the validity of HAI and then applied HAI to 2015 KBO data in which HAI is analyzed statistically with respect to different teams, ages, and positions. Moreover, the linear relationship between KBO batter's HAI and their annual salary is discussed. Grouping 46 KBO batters based on confidence region of the regression model for annual salary, we also statistically investigate batter's annual salary in these groups with respect to several factors.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.34 no.3
• /
• pp.849-858
• /
• 2014

The mount of antecedent 5-day rainfall (P5) is usually used to determine the antecedent soil moisture condition for estimating effective rainfall using the NRCS-CN method. In order to re-establish the threshold of P5 considering basin characteristics, this study investigated the sensitivity of the threshold of P5 to effective rainfall by comparing the corresponding observed direct runoff. The overall results indicate that the direct runoff estimated using the re-establihed threshold of P5 has smaller mean error (RMSE of 27.3 mm) than those using the conventional threshold (RMSE of 35.2 mm). In addition, after evaluating the effectiveness of threshold of P5 using the improvement index, the threshold re-established in this study improved the ability to estimate the direct runoff by 30% on average. This study also suggested to employ regression models using topographic indices to re-establish the threshold for ungauged basins. When using the re-established threshold from the regression model, the RMSE decreased ranging from 0.4 mm to 15.1 mm and the efficiency index of Nash and Sutcliffe increased up to 0.33.

• International Journal of Highway Engineering
• /
• v.10 no.4
• /
• pp.151-159
• /
• 2008

Performances of retroreflectivity vary place to place, according to traffic volumes and time lengths after striping, depending on pavement marking materials and colors. The present paper uses the nation wide data of retroreflectivity, which has been collected from freeways and then tries to develop the regression curve setting traffic volume and service life as independent variables and retroreflectivities as dependent variables. The DB system includes two year's measurement in $2005{\sim}2006$ over Korean freeway pavement marking at an interval of three months for the period. The mobile measurement system, a laserlux, was employed for the purpose. The DB has provided a lot of information about materials and performance of the specific pavement marking such as geometric features, traffic volumes, material characteristics and the installation date. This study provides the comparison of pavement marking performances under diversified conditions. Based on accumulated pavement marking performances, this study provides performance curves based on the diversified factors. The goal of the retroreflectivity modeling is to develop equations that can be used to estimate an average retroreflectivity of pavement markings as a function time since application and traffic volume. After representing the variation of retroreflectivities and estimating regression curves by linear, exponential, logarithmic and power function, the regression curve which had the highest coefficient of determination and the value similar to the last field measurement was regarded as the retroreflectivity decay model. As a result of verification, the decay model showed the signification within the 90% confidence level and especially showed the clear relation with field data according to increase of cumulative vehicle exposure. Accordingly, these models can be used to determine service lives, retroreflectivity degradation rates, and retroreflectivity of new markings.

• Journal of the Korean Institute of Traditional Landscape Architecture
• /
• v.29 no.3
• /
• pp.147-153
• /
• 2011

What kinds of environmental adaptation did people during the Chosun era take when climate change due to drought was perceived and natural disasters occurred? Research hypotheses is 'More environmental improvement projects were conducted when climate disasters were experienced than rituals for rain calling. In this study, frequency samples for Rituals for Rain (RR, 祈雨祭) defined as abnormal climate of drought, Climate Disaster(CD) and Environmental Improvement Projects(EIP) were extracted from "Chosunwangjoshillok(朝鮮王朝實錄)". The analysis among RR, CD, and EIP were studied for a regression model. Research hypothesis was statistically tested. RR took placed the order of Sejong(世宗), Sukjong(肅宗) and Youngjo(英祖), while climate disasters were extracted under the Jungjong(中宗), Sungjong(成宗) and Taijong(太宗). EIP were most active under Youngjo(英祖), Sungjong(成宗) and Taijong(太宗). During the former part of the Chosun dynasty, abnormal climate was more seriously considered than climate disasters, while the opposite pattern was shown during the latter part. In a regression analysis between EIP and CD, the equation of EIP = 0.632CD was determined. As a test result, the study hypothesis the entire Chosun dynasty was dismissed. However, it is possible to statistically support that more EIP were conducted in order to adapt to the climate change during the latter part of the dynasty when climate disaster were experienced.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.4
• /
• pp.785-789
• /
• 2021

Paprika productivity is different even in the same quality greenhouse and in the same region. These differences are known to due to differences in various environmental factors. This study was conducted to investigate the difference in the level of various environmental factors between high-productivity (HPF) and low-productivity (LPF) greenhouses. The largest difference between the two greenhouses in the daily or weekly average values of major environmental factors was the CO₂ concentration, but the LPF was higher than the HPF, so it was not determined as a factor for the difference in productivity. Correlation analysis among 14 environmental factors showed a high correlation among irradiation or related factors in moisture. The regression coefficients of the linear regression model between vapor pressure deficit and relative humidity were -0.0202kpa in HPF and -0.0262kpa in LPF. In particular, in February and March, the vapor pressure deficit in LPF was 1.5kpa or more, and the cumulative vapor pressure deficit compared to the cumulative irradiation at the early period of cultivation increased rapidly. The reason for the low productivity in LPF is thought to be that the plant was affected by moisture stress due to high vapor pressure deficit and transpiration under low irradiation conditions in the early period of cultivation and in winter.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.1
• /
• pp.9-18
• /
• 2023

Predicting the compressive strength of high-performance concrete (HPC) is challenging because of the use of additional cementitious materials; thus, the development of improved predictive models is essential. The purpose of this study was to develop an HPC compressive-strength prediction model using an ensemble machine-learning method of combined bagging and stacking techniques. The result is a new ensemble technique that integrates the existing ensemble methods of bagging and stacking to solve the problems of a single machine-learning model and improve the prediction performance of the model. The nonlinear regression, support vector machine, artificial neural network, and Gaussian process regression approaches were used as single machine-learning methods and bagging and stacking techniques as ensemble machine-learning methods. As a result, the model of the proposed method showed improved accuracy results compared with single machine-learning models, an individual bagging technique model, and a stacking technique model. This was confirmed through a comparison of four representative performance indicators, verifying the effectiveness of the method.

• Economic and Environmental Geology
• /
• v.48 no.2
• /
• pp.147-160
• /
• 2015

In 2014, on 31 August and 1 September, the emissions of $CH_4$, $CO_2$, and $O_2$ gases were measured six times using the closed chamber method from exposed tidal flat sediments in the same position relative to the low point of the tidal cycle in the Eoeun-ri, Taean-gun, on the Mid-western Coast of Korea. The concentrations of $CH_4$ in the air sample collected in the chamber were measured using gas chromatography with an EG analyzer, model GS-23, within 6 hours of collection, and the other gases were measured in real time using a multi-gas monitor. The gas emission fluxes (source (+), and sink (-)) were calculated from a simple linear regression analysis of the changes in the concentrations over time. In order to see the surrounding parameters (water content, temperature, total organic carbon, average mean size of sediments, and the temperature of the inner chamber) were measured at the study site. On the first day, across three measurements during 5 hours 20 minutes, the observed $CO_2$ flux absorption was -137.00 to $-81.73mg/m^2/hr$, and the $O_2$ absorption, measured simultaneously, was -0.03 to $0.00mg/m^2/hr$. On the second day using an identical number of measurements, the $CO_2$ absorption was -20.43 to $-2.11mg/m^2/hr$, and the $O_2$ absorption -0.18 to $-0.14mg/m^2/hr$. The $CH_4$ absorption before low tide was $-0.02mg/m^2/hr$ (first day, Pearson correlation coefficient using the SPSS statistical analysis is -0.555(n=5, p=0.332, pronounced negative linear relationship)), and $-0.15mg/m^2/hr$ (second day, -0.915(n=5, p=0.030, strong negative linear relationship)) on both measurement days. The emitted flux after low tide on both measurement days reached a minimum of $+0.00mg/m^2/hr$ (+0.713(n=5, p=0.176, linear relationship which can be almost ignored)), and a maximum of $+0.03mg/m^2/hr$ (+0.194(n=5, p=0.754, weak positive linear relationship)) after low tide. However, the absolute values of the $CH_4$ fluxes were analyzed at different times. These results suggest that rate for $CH_4$ fluxes, even the same time and area, were influenced by changes in the tidal cycle characteristics of surface sediments for understanding their correlation with these gas emissions, and surrounding parameters such as physiochemical sediments conditions.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.127-146
• /
• 2022

Recently, as word embedding has shown excellent performance in various tasks of deep learning-based natural language processing, researches on the advancement and application of word, sentence, and document embedding are being actively conducted. Among them, cross-language transfer, which enables semantic exchange between different languages, is growing simultaneously with the development of embedding models. Academia's interests in vector alignment are growing with the expectation that it can be applied to various embedding-based analysis. In particular, vector alignment is expected to be applied to mapping between specialized domains and generalized domains. In other words, it is expected that it will be possible to map the vocabulary of specialized fields such as R&D, medicine, and law into the space of the pre-trained language model learned with huge volume of general-purpose documents, or provide a clue for mapping vocabulary between mutually different specialized fields. However, since linear-based vector alignment which has been mainly studied in academia basically assumes statistical linearity, it tends to simplify the vector space. This essentially assumes that different types of vector spaces are geometrically similar, which yields a limitation that it causes inevitable distortion in the alignment process. To overcome this limitation, we propose a deep learning-based vector alignment methodology that effectively learns the nonlinearity of data. The proposed methodology consists of sequential learning of a skip-connected autoencoder and a regression model to align the specialized word embedding expressed in each space to the general embedding space. Finally, through the inference of the two trained models, the specialized vocabulary can be aligned in the general space. To verify the performance of the proposed methodology, an experiment was performed on a total of 77,578 documents in the field of 'health care' among national R&D tasks performed from 2011 to 2020. As a result, it was confirmed that the proposed methodology showed superior performance in terms of cosine similarity compared to the existing linear vector alignment.