• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.1
• /
• pp.317-326
• /
• 2021

This study examines factors influencing the corporate identity (CI) of pharmacy retail stores (PRSs) in Thailand as a means to construct a unique corporate identity to represent their strengths and image abroad. An exploratory sequential design was applied. A pilot study involving interviews with four pharmacists was conducted to obtain the variables, and then a questionnaire was designed and administered to 392 respondents. The collected data was analyzed to examine correlations via descriptive analysis, Pearson's correlation, and multiple linear regression (MLR). The results show that wisdom and pharmaceutical care explain 44% of the variance in defining Thai PRSs identity. There is a 61% chance that Thai PRSs should consider wisdom as the most important factor. These findings provide useful insights for pharmacists, pharmaceutical entrepreneurs, and owners of PRSs to enhance competitiveness by devising strategies to create their corporate identity. For the qualitative analysis, Thai PRSs should encourage pharmacists to gain real working experience to develop their wisdom, experience, and skills. Besides, PRSs that build identities as knowledgeable, sincere, and compassionate health providers can successfully expand their operations to other ASEAN countries, as end-consumers will be confident in the reliability of their services.

• Journal of Distribution Science
• /
• v.20 no.5
• /
• pp.105-117
• /
• 2022

Purpose: This research investigates ready-to-drink (RTD) coffee. Although the RTD coffee market is growing competitively, few studies have examined behavioral re-intention or repurchase intention in the context of this industry. Therefore, the objective of this study was to explore factors affecting the behavioral re-intention to purchase RTD coffee. Research design, data and methodology: Using the theory of planned behavior (TPB) as the underpinning theoretical framework, this study hypothesized that behavioral re-intention to purchase RTD coffee is influenced by the variables of the TPB and additional variables. A mixed-method research design was applied, starting with qualitative in-depth interviews and followed by a quantitative method. Data were collected using an online survey of coffee lovers. Multiple linear regression (MLR) was used to assess the hypothesized relationships in the proposed conceptual framework. Results: The results reveal that content sensory attribute beliefs are the strongest positive predictor of behavioral re-intention in Thailand, followed by perceived utilitarian value. In contrast, price signaling was negatively related to behavioral re-intention. Conclusions: The findings can help food and beverage companies to develop new coffee product lines to gain more market share, create integrated marketing communications to build brand awareness, and manage distribution channels and the supply chain.

• Geomechanics and Engineering
• /
• v.37 no.6
• /
• pp.539-554
• /
• 2024

This study explores development of prediction model for seismic site classification through the integration of machine learning techniques with horizontal-to-vertical spectral ratio (HVSR) methodologies. To improve model accuracy, the research employs outlier detection methods and, synthetic minority over-sampling technique (SMOTE) for data balance, and evaluates using seven machine learning models using seismic data from KiK-net. Notably, light gradient boosting method (LGBM), gradient boosting, and decision tree models exhibit improved performance when coupled with SMOTE, while Multiple linear regression (MLR) and Support vector machine (SVM) models show reduced efficacy. Outlier detection techniques significantly enhance accuracy, particularly for LGBM, gradient boosting, and voting boosting. The ensemble of LGBM with the isolation forest and SMOTE achieves the highest accuracy of 0.91, with LGBM and local outlier factor yielding the highest F1-score of 0.79. Consistently outperforming other models, LGBM proves most efficient for seismic site classification when supported by appropriate preprocessing procedures. These findings show the significance of outlier detection and data balancing for precise seismic soil classification prediction, offering insights and highlighting the potential of machine learning in optimizing site classification accuracy.

• Journal of Radiation Industry
• /
• v.17 no.4
• /
• pp.551-557
• /
• 2023

In recent years, radiation has become a socially important issue, increasing the need for accurate prediction of radiation levels. In this study, machine learning-based models such as Multiple Linear Regression (MLR), Random Forest (RF), XGBoost, and LightGBM, which predict the dose rate by time(nSv h^-1) by selecting only important variables, were used, and the correlation between temperature, humidity, cumulative precipitation, wind direction, wind speed, local air pressure, sea pressure, solar radiation, and radiation dose rate (nSv h^-1) was analyzed by collecting weather data and radiation dose rate for about 6 months in Jangseong, Jeollanam-do. As a result of the evaluation based on the RMSE (Root Mean Squared Error) and R-Squared (R-Squared coefficient of determination) scores, the RMSE of the XGBoost model was 22.92 and the R-Squared was 0.73, showing the best performance among the models used. As a result of optimizing hyperparameters of all models using the GridSearch method and comparing them by adding variables inside the measuring instrument, it was confirmed that the performance improved to 2.39 for RMSE and 0.99 for R-Squared in both XGBoost and LightGBM.

• Journal of the Korean Chemical Society
• /
• v.56 no.2
• /
• pp.236-245
• /
• 2012

The main mycobacterial infection in human is tuberculosis caused by Mycobacterium tuberculosis. Tuberculosis is the leading infectious cause of death in the world. Therefore there is continuing and compelling need for new and improved treatment for tuberculosis. The entire logic towards design of new compounds containing 4-hydroxy-N'-(1,3-thiazoldin- 2-yldene)benzohydrazide moiety is basically for superior antimycobacterial activity. The recent advances in QSAR and computer science have provided a systematic approach to design a structure of any compound and further, the biological activity of the compound can be predicted before synthesis. The 3D-QSAR studies for the set of 4-hydroxy-N'-(1,3-thiazoldin- 2-yldene)benzohydrazide and their derivatives were carried out by using V-life MDS (3.50). The various statistical methods such as Multiple Linear Regression (MLR), Partial Least Square Regression (PLSR), Principle Component Regression(PCR) and K nearest neighbour (kNN) were used. The kNN showed good results having cross validated $r^2$ 0.9319, $r^2$ for external test set 0.8561 and standard error of estimate 0.2195. The docking studies were carried out by using Schrodinger GLIDE module which resulted in good docking score in comparison with the standard isoniazid. The designed compounds were further subjected for synthesis and biological evaluation. Antitubercular evaluation of these compounds showed that (4.a), (4.d) and (4.g) found as potent inhibitor of H37RV.

• Journal of KIISE:Databases
• /
• v.34 no.6
• /
• pp.483-492
• /
• 2007

Time series of conventional prediction techniques uses the model which is generated from the training step. This model is applied to new input data without any change. If this model is applied directly to stream data, the rate of prediction accuracy will be decreased. This paper proposes an stream data prediction technique using sliding window and regression. This technique considers the characteristic of time series which may be changed over time. It is composed of two steps. The first step executes a fractional process for applying input data to the regression model. The second step updates the model by using its information as new data. Additionally, the model is maintained by only recent data in a queue. This approach has the following two advantages. It maintains the minimum information of the model by using a matrix, so space complexity is reduced. Moreover, it prevents the increment of error rate by updating the model over time. Accuracy rate of the proposed method is measured by RME(Relative Mean Error) and RMSE(Root Mean Square Error). The results of stream data prediction experiment are performed by the proposed technique IMQR(Incremental Multiple Quadratic Regression) is more efficient than those of MLR(Multiple Linear Regression) and SVR(Support Vector Regression).

• Journal of Bio-Environment Control
• /
• v.32 no.4
• /
• pp.384-395
• /
• 2023

Since transpiration plays a key role in optimal irrigation management, knowledge of the irrigation demand of crops like tomatoes, which are highly susceptible to water stress, is necessary. One way to determine irrigation demand is to measure transpiration, which is affected by environmental factor or growth stage. This study aimed to estimate the transpiration amount of tomatoes and find a suitable model using mathematical and deep learning models using minute-by-minute data. Pearson correlation revealed that observed environmental variables significantly correlate with crop transpiration. Inside air temperature and outside radiation positively correlated with transpiration, while humidity showed a negative correlation. Multiple Linear Regression (MLR), Polynomial Regression model, Artificial Neural Network (ANN), Long short-term Memory (LSTM), and Gated Recurrent Unit (GRU) models were built and compared their accuracies. All models showed potential in estimating transpiration with R² values ranging from 0.770 to 0.948 and RMSE of 0.495 mm/min to 1.038 mm/min in the test dataset. Deep learning models outperformed the mathematical models; the GRU demonstrated the best performance in the test data with 0.948 R² and 0.495 mm/min RMSE. The LSTM and ANN closely followed with R² values of 0.946 and 0.944, respectively, and RMSE of 0.504 m/min and 0.511, respectively. The GRU model exhibited superior performance in short-term forecasts while LSTM for long-term but requires verification using a large dataset. Compared to the FAO56 Penman-Monteith (PM) equation, PM has a lower RMSE of 0.598 mm/min than MLR and Polynomial models degrees 2 and 3 but performed least among all models in capturing variability in transpiration. Therefore, this study recommended GRU and LSTM models for short-term estimation of tomato transpiration in greenhouses.

• Analytical Science and Technology
• /
• v.24 no.6
• /
• pp.544-555
• /
• 2011

The acute toxicity in the rainbow trout (Oncorhynchus mykiss) was analyzed and predicted using quantitative structure-activity relationships (QSAR). The aquatic toxicity, 96h $LC_{50}$ (median lethal concentration) of 275 organic pesticides, was obtained from EU-funded project DEMETRA. Prediction models were derived from 558 2D molecular descriptors, calculated in PreADMET. The linear (multiple linear regression) and nonlinear (support vector machine and artificial neural network) learning methods were optimized by taking into account the statistical parameters between the experimental and predicted p$LC_{50}$. After preprocessing, population based forward selection were used to select the best subsets of descriptors in the learning methods including 5-fold cross-validation procedure. The support vector machine model was used as the best model ($R^2_{CV}$=0.677, RMSECV=0.887, MSECV=0.674) and also correctly classified 87% for the training set according to EU regulation criteria. The MLR model could describe the structural characteristics of toxic chemicals and interaction with lipid membrane of fish. All the developed models were validated by 5 fold cross-validation and Y-scrambling test.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1622-1622
• /
• 2001

The purpose of this research was to develop a the reflection technique with near infrared (NIR) radiation for estimating physico-chemical properties in compost. The composts (cattle, pig, chicken and waste composts) were air dried and then ground to pass through a 0.5 or 2mm sieve for the physico-chemical properties and spectroscopic determinations. The physico-chemical properties of compost were shown high values ; moisture(30-60%), T-N(0.8-2.9%), organic matter(29-89%), pH(5.89-9.60) K$_2$O(0.27-5.66%), P2O$\sub$5/(0.07-2.62%), CaO(0.03-4.80%), MgO(0.09-1.56%), NaCl(0.01-1.13%), EC(1.41-13.76dS/m). Generally, we should select a simple calibration and prediction method for determining physico-chemical properties in compost under similar accuracy and precision of prediction. It should be remembered that the NIRS approach will never replace the traditional methods. However, NIRS technique may be an effective method for rapid and nondestructive measurements of a large number of compost samples. Near infrared reflectance spectra of composts was obtained by Infra Alyzer 500 scanning spectrophotometer at 2-nm intervals from 1100 to 2500nm. Multiple linear regression(MLR) or partial least square regression (PLSR) was used to evaluate a NIRS method for the rapid and nondestructive determination of physico-chemical properties and humic acid contents in composts. The standard error of prediction(SEP) for finely sized sample(<0.5mm) and coarsely sized sample(<2mm) did not show much difference. The NIR instrument of filter type showed the same accuracy of the monochromator scanning type to estimate the compost properties. The results summarized that NIR spectroscopy can be used as a routine testing method to determine quantitatively the OM, moisture, T-N, color, pH, cation content in the compost samples nondestructively.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.54 no.1
• /
• pp.29-35
• /
• 2009

The aim of the present study was to update the calibration that is used for the measurement of the total nitrogen content in the rice plant samples by using the visible and near infrared spectrum. Before the equation merge, correlation coefficient of calibration equation for nitrogen content on each rice parts was 0.945 (Leaf), 0.928 (Stem), and 0.864 (Whole plant), respectively. In the calibration models created by each part in the rice plant under the various regression method, the calibration model for the leaf was recorded with relatively high accuracy. Among of those, the calibration equation developed by Partial least squares (PLS) method was more accurate than the Multiple linear regression (MLR) method. The calibration equation was sensitive based on variety and location variations. However, we have merged and enlarged various of the samples that made not only to measure the nitrogen content more accurately, but also later sampling populations became more diversified. After merging, $R^2$ value becomes more accurate and significantly to 0.950 (L.), 0.974 (S.), 0.940 (W.). Also, after removal of outlier, R2 values increased into 0.998, 0.995, and 0.997. In view of the results so far achieved, Standard error of prediction (SEP) and SEP (C) were reduced in the stem and whole plant. Biases were reduced in the leaf, stem as well as whole plant. Slopes were high in the stem. Standard deviation reduced in the stem but $R^2$ was high in the stem and whole plant. Result was indicated that calibration equation make update, and updating robust calibration equation from merge function and multi-variate calibration.