• Journal of Applied Biological Chemistry
• /
• v.62 no.2
• /
• pp.129-135
• /
• 2019

Salvia miltiorrhiza Radix is cultivated in Korea and China and is traditionally used to treat cardiovascular diseases. In this study, we developed and validated a quantitative analysis method for S. miltiorrhiza Radix using high-performance liquid chromatography (HPLC). Identification was performed using ultra performance liquid chromatography-tandem mass spectrometry. For quantitative analysis, we used seven marker compounds. Separation conditions for HPLC were optimized using an ODS column with gradient conditions of 1% formic acid in distilled water and 1% formic acid in acetonitrile, with a flow rate of 0.8 mL/min and a detection wavelength of 280 nm. This method showed good linearity ($R^2=0.9998$), precision (relative standard deviation ${\leq}3.3%$), accuracy (recovery of 94.16-102.89%), limit of detection ($7.53{\mu}g/mL$), and limit of quantification ($23.71{\mu}g/mL$). This approach successfully quantified marker compounds in S. miltiorrhiza Radix. The individual marker compounds were identified by comparing the molecular masses and retention times with does standard compounds. Marker compound contents of S. miltiorrhiza Radix were investigated with different cultivation regions. Seven marker compounds were detected and quantified in all samples. Among them, salvianolic acid B showed the highest contents and it ranged from 4.13 to 7.15%. The salvianolic acid B content (7.15%) of marker compound was the highest in Bonghwa, and the tanshinone IIA content (1.90%) was the highest in Pohang. The results of marker compounds and developed method were intended to provide a favorable reference for the study of S. miltiorrhiza Radix from different regions of Korea.

• The Korean Journal of Applied Statistics
• /
• v.34 no.2
• /
• pp.153-166
• /
• 2021

A main goal of pharmacogenomics studies is to predict individual's drug responsiveness based on high dimensional genetic variables. Due to a large number of variables, feature selection is required in order to reduce the number of variables. The selected features are used to construct a predictive model using machine learning algorithms. In the present study, we applied several hybrid feature selection methods such as combinations of logistic regression, ReliefF, TurF, random forest, and LASSO to a next generation sequencing data set of 400 epilepsy patients. We then applied the selected features to machine learning methods including random forest, gradient boosting, and support vector machine as well as a stacking ensemble method. Our results showed that the stacking model with a hybrid feature selection of random forest and ReliefF performs better than with other combinations of approaches. Based on a 5-fold cross validation partition, the mean test accuracy value of the best model was 0.727 and the mean test AUC value of the best model was 0.761. It also appeared that the stacking models outperform than single machine learning predictive models when using the same selected features.

• The Journal of Information Systems
• /
• v.30 no.4
• /
• pp.21-42
• /
• 2021

Purpose In order for companies to continue to grow, they should properly manage human resources, which are the core of corporate competitiveness. Employee turnover means the loss of talent in the workforce. When an employee voluntarily leaves his or her company, it will lose hiring and training cost and lead to the withdrawal of key personnel and new costs to train a new employee. From an employee's viewpoint, moving to another company is also risky because it can be time consuming and costly. Therefore, in order to reduce the social and economic costs caused by employee turnover, it is necessary to accurately predict employee turnover intention, identify the factors affecting employee turnover, and manage them appropriately in the company. Design/methodology/approach Prior studies have mainly used logistic regression and decision trees, which have explanatory power but poor predictive accuracy. In order to develop a more accurate prediction model, XGBoost is proposed as the classification technique. Then, to compensate for the lack of explainability, SHAP, one of the XAI techniques, is applied. As a result, the prediction accuracy of the proposed model is improved compared to the conventional methods such as LOGIT and Decision Trees. By applying SHAP to the proposed model, the factors affecting the overall employee turnover intention as well as a specific sample's turnover intention are identified. Findings Experimental results show that the prediction accuracy of XGBoost is superior to that of logistic regression and decision trees. Using SHAP, we find that jobseeking, annuity, eng_test, comm_temp, seti_dev, seti_money, equl_ablt, and sati_safe significantly affect overall employee turnover intention. In addition, it is confirmed that the factors affecting an individual's turnover intention are more diverse. Our research findings imply that companies should adopt a personalized approach for each employee in order to effectively prevent his or her turnover.

• Journal of Information Technology Services
• /
• v.21 no.3
• /
• pp.63-72
• /
• 2022

The extinction crisis of local cities, caused by a population density increase phenomenon in capital regions, directly causes the increase of vacant houses in local cities. According to population and housing census, Gunsan-si has continuously shown increasing trend of vacant houses during 2015 to 2019. In particular, since Gunsan-si is the city which suffers from doughnut effect and industrial decline, problems regrading to vacant house seems to exacerbate. This study aims to provide a foundation of a system which can predict and deal with the building that has high risk of becoming vacant house through implementing a data driven vacant house prediction machine learning model. Methodologically, this study analyzes three types of machine learning model by differing the data components. First model is trained based on building register, individual declared land value, house price and socioeconomic data and second model is trained with the same data as first model but with additional POI(Point of Interest) data. Finally, third model is trained with same data as the second model but with excluding water usage and electricity usage data. As a result, second model shows the best performance based on F1-score. Random Forest, Gradient Boosting Machine, XGBoost and LightGBM which are tree ensemble series, show the best performance as a whole. Additionally, the complexity of the model can be reduced through eliminating independent variables that have correlation coefficient between the variables and vacant house status lower than the 0.1 based on absolute value. Finally, this study suggests XGBoost and LightGBM based machine learning model, which can handle missing values, as final vacant house prediction model.

• International Journal of Computer Science & Network Security
• /
• v.24 no.2
• /
• pp.101-112
• /
• 2024

Telugu language is considered as fourth most used language in India especially in the regions of Andhra Pradesh, Telangana, Karnataka etc. In international recognized countries also, Telugu is widely growing spoken language. This language comprises of different dependent and independent vowels, consonants and digits. In this aspect, the enhancement of Telugu Handwritten Character Recognition (HCR) has not been propagated. HCR is a neural network technique of converting a documented image to edited text one which can be used for many other applications. This reduces time and effort without starting over from the beginning every time. In this work, a Unicode based Handwritten Character Recognition(U-HCR) is developed for translating the handwritten Telugu characters into English language. With the use of Centre of Gravity (CG) in our model we can easily divide a compound character into individual character with the help of Unicode values. For training this model, we have used both online and offline Telugu character datasets. To extract the features in the scanned image we used convolutional neural network along with Machine Learning classifiers like Random Forest and Support Vector Machine. Stochastic Gradient Descent (SGD), Root Mean Square Propagation (RMS-P) and Adaptative Moment Estimation (ADAM)optimizers are used in this work to enhance the performance of U-HCR and to reduce the loss function value. This loss value reduction can be possible with optimizers by using CNN. In both online and offline datasets, proposed model showed promising results by maintaining the accuracies with 90.28% for SGD, 96.97% for RMS-P and 93.57% for ADAM respectively.

• Progress in Medical Physics
• /
• v.19 no.4
• /
• pp.269-275
• /
• 2008

Although MR imaging is generally applicable to depict knee joint deterioration it, is sometimes occurred to mis-read and mis-diagnose the common knee joint diseases. In this study, we employed magnetization transfer ratio (MTR) method to improve the diagnosis of the various knee joint diseases. Spin-echo (SE) T2-weighted images (TR/TE 3,400-3,500/90-100 ms) were obtained in seven cases of knee joint deterioration, FSE T2-weighted images (TR/TE 4,500-5,000/100-108 ms) were obtained in seven cases of knee joint deterioration, gradient-echo (GRE) T2-weighted images (TR/TE 9/4.56/$50^{\circ}$ flip angle, NEX 1) were obtained in 3 cases of knee joint deterioration, In six cases of knee joint deterioration, fat suppression was performed using a T2-weighted short T1/tau inverse recovery (STIR) sequence (TR/TE =2,894-3,215 ms/70 ms, NEX 3, ETL 9). Calculation of MTR for individual pixels was performed on registration of unsaturated and saturated images. After processing to make MTR images, the images were displayed in gray color. For improving diagnosis, three-dimensional isotropic volume images, the MR tristimulus color mapping and the MTR map was employed. MTR images showed diagnostic images quality to assess the patients' pathologies. The intensity difference between MTR images and conventional MRI was seen on the color bar. The profile graph on MTR imaging effect showed a quantitative measure of the relative decrease in signal intensity due to the MT pulse. To diagnose the pathologies of the knee joint, the profile graph data was shown on the image as a small cross. The present study indicated that MTR images in the knee joint were feasible. Investigation of physical change on MTR imaging enables to provide us more insight in the physical and technical basis of MTR imaging. MTR images could be useful for rapid assessment of diseases that we examine unambiguous contrast in MT images of knee disorder patients.

• Asian-Australasian Journal of Animal Sciences
• /
• v.27 no.11
• /
• pp.1562-1570
• /
• 2014

The effects of Lactobacillus mucosae (L. mucosae), a potential direct fed microbial previously isolated from the rumen of Korean native goat, on the rumen fermentation profile of brewers grain were evaluated. Fermentation was conducted in serum bottles each containing 1% dry matter (DM) of the test substrate and either no L. mucosae (control), 1% 24 h broth culture of L. mucosae (T1), or 1% inoculation with the cell-free culture supernatant (T2). Each serum bottle was filled anaerobically with 100 mL of buffered rumen fluid and sealed prior to incubation for 0, 6, 12, 24, and 48 h from which fermentation parameters were monitored and the microbial diversity was evaluated. The results revealed that T1 had higher total gas production (65.00 mL) than the control (61.33 mL) and T2 (62.00 mL) (p<0.05) at 48 h. Consequently, T1 had significantly lower pH values (p<0.05) than the other groups at 48 h. Ammonia nitrogen ($NH_3$-N), individual and total volatile fatty acids (VFA) concentration and acetate:propionate ratio were higher in T1 and T2 than the control, but T1 and T2 were comparable for these parameters. Total methane ($CH_4$) production and carbon dioxide ($CO_2$) were highest in T1. The percent DM and organic matter digestibilities were comparable between all groups at all times of incubation. The total bacterial population was significantly higher in T1 (p<0.05) at 24 h, but then decreased to levels comparable to the control and T2 at 48 h. The denaturing gradient gel electrophoresis profile of the total bacterial 16s rRNA showed higher similarity between T1 and T2 at 24 h and between the control and T1 at 48 h. Overall, these results suggest that addition of L. mucosae and cell-free supernatant during the in vitro fermentation of dried brewers grain increases the VFA production, but has no effect on digestibility. The addition of L. mucosae can also increase the total bacterial population, but has no significant effect on the total microbial diversity. However, inoculation of the bacterium may increase $CH_4$ and $CO_2$ in vitro.

• Journal of The Korean Astronomical Society
• /
• v.50 no.3
• /
• pp.79-92
• /
• 2017

We present a BV I optical photometric study of the old open cluster Ruprecht 6 using the data obtained with the SMARTS 1.0 m telescope at the CTIO, Chile. Its color-magnitude diagrams show the clear existence of the main-sequence stars, whose turn-off point is located around $V{\approx}18.45mag$ and $B-V{\approx}0.85mag$. Three red clump (RC) stars are identified at V = 16.00 mag, I = 14.41 mag and B - V = 1.35 mag. From the mean $K_s-band$ magnitude of RC stars ($K_s=12.39{\pm}0.21mag$) in Ruprecht 6 from 2MASS photometry and the known absolute magnitudes of the RC stars ($M_{K_S}=-1.595{\pm}0.025mag$), we obtain the distance modulus to Ruprecht 6 of $(m-M)_0=13.84{\pm}0.21mag$ ($d=5.86{\pm}0.60kpc$). From the ($J-K_s$) and (B - V ) colors of the RC stars, comparison of the (B - V ) and (V - I) colors of the bright stars in Ruprecht 6 with those of the intrinsic colors of dwarf and giant stars, and the PARSEC isochrone fittings, we derive the reddening values of E(B - V ) = 0.42 mag and E(V - I) = 0.60 mag. Using the PARSEC isochrone fittings onto the color-magnitude diagrams, we estimate the age and metallicity to be: $log(t)=9.50{\pm}0.10(t=3.16{\pm}0.82Gyr)$ and $[Fe/H]=-0.42{\pm}0.04dex$. We present the Galactocentric radial metallicity gradient analysis for old (age > 1 Gyr) open clusters of the Dias et al. catalog, which likely follow a single relation of $[Fe/H]=(-0.034{\pm}0.007)R_{GC}+(0.190{\pm}0.080)$ (rms = 0.201) for the whole radial range or a dual relation of $[Fe/H]=(-0.077{\pm}0.017)R_{GC}+(0.609{\pm}0.161)$ (rms = 0.152) and constant ([Fe/H] ~ -0.3 dex) value, inside and outside of RGC ~ 12 kpc, respectively. The metallicity and Galactocentric radius ($13.28{\pm}0.54kpc$) of Ruprecht 6 obtained in this study seem to be consistent with both of the relations.

• Journal of Biosystems Engineering
• /
• v.42 no.2
• /
• pp.86-97
• /
• 2017

Purpose: This study aims to evaluate the applicability of a tractor-baler system equipped with a newly developed round baler by conducting stability analyses via static-state mathematical simulations and verification experiments for the tractor equipped with a loader. Methods: The centers of gravity of the tractor and baler were calculated to analyze the transverse overturning of the system. This overturning of the system was analyzed by applying mathematical equations presented in previous research and comparing the results with those obtained by the newly developed mathematical simulation. For the case of the tractor equipped with a loader, mathematical simulation results and experimental values from verification experiments were compared and verified. Results: The center of gravity of the system became lower after the baler was attached to the tractor and the angle of transverse overturning of the system steadily increased or decreased as the deflection angle increased or decreased between $0^{\circ}$ and $180^{\circ}$ on the same gradient. In the results of the simulations performed by applying mathematical equations from previous research, right transverse overturning occurred when the tilt angle was at least $19.5^{\circ}$ and the range of deflection angles was from $82^{\circ}$ to $262^{\circ}$ in counter clockwise. Additionally, left transverse overturning also occurred at tilt angles of at least $19.5^{\circ}$ and the range of deflection angles was from $259^{\circ}$ to $79^{\circ}$ in counter clockwise. Under the $0^{\circ}$ deflection angle condition, in simulations of the tractor equipped with a loader, transverse overturning occurred at $17.9^{\circ}$, which is a 2.3% change from the results of the verification experiment ($17.5^{\circ}$). The simulations applied the center of gravity and the correlations between the tilt angles, formed by individual wheel ground contact points excluding wheel radius and hinge point height, which cannot be easily measured, for the convenient use of mathematical equations. The results indicated that both left and right transverse overturning occurred at $19.5^{\circ}$. Conclusions: The transverse overturning stability evaluation of the system, conducted via mathematical equation modeling, was stable enough to replace the mathematical equations proposed by previous researchers. The verification experiments and their results indicated that the system is workable at $12^{\circ}$, which is the tolerance limit for agricultural machines on the sloped lands in South Korea, and $15^{\circ}$, which is the tolerance limit for agricultural machines on the sloped grasslands of hay in Japan.

• Journal of Animal Science and Technology
• /
• v.45 no.5
• /
• pp.731-738
• /
• 2003

The technique known as proteomics is useful for characterizing the protein expression pattern of a particular tissue or cell type as well as quantitatively identifying differences in the levels of individual proteins. In present study, we carried out the comparative expression patterns of white and red muscles. We used the two-dimensional electrophoresis(2-DE) for analyzing the protein expression. Proteins isolated from porcine white and red muscles were separated by 12% poly-acrylamide gel and then were detected by coomassie blue and silver staining. More than 600 protein spots were detected on each 2-DE gel. By visual analysis of the stained gel, five proteins were identified to be differentially expressed in the white vs red muscle. By database searching based on the molecular weights and pI(isoelectric point) of the five proteins, three of them were found to be most close to troponin I, T and myoglobin. However, further researche is needed for identification and functional analysis of the unidentified proteins. In conclusion, we found five proteins, which are differentially expressed in the white vs red muscle. The functional analysis of the differentially expressed proteins will provide valuable information on biochemical characteristics of the muscle type.