• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.481-481
• /
• 2000

We investigated the structure-activity relationships on use of multi-layer neural networks. The relationships are techniques required in developments of medicines. Since many kinds of observations might be adopted on the techniques, we discussed some points between the observations and the properties of multi-layer neural networks. In the structure-activity relationships, an important property is not that standard deviations are nearly equal to zero for observed physiological activity, but prediction ability for unknown medicines. Since we adopted non-linear approximation, the function to represent the activity can be defined by observations; therefore, we believe that the standard deviations have not significance. The function was examined by "leave-one-out" method, which was originally introduced for the multi-regression analysis. In the linear approximation, the examination is significance, however, we believe that the method is inappropriate in case of nonlinear fitting as neural networks; therefore, we derived a new index fer the relationships from the differential of information propagation in the neural network. By using the index, we discussed physiological activity of an anti-cancer medicine, Mitomycine derivatives. The neuro-computing suggests that there is no direction to extend the anti-cancer activity of Mitomycine, which is close to the trend of anticancer developing.

• Journal of the Korean Wood Science and Technology
• /
• v.21 no.4
• /
• pp.64-72
• /
• 1993

Mechanical properties of 15 mm thick, three-layer particleboard were studied by varying resin content, specific gravity, mat moisture content, pressing time and pressing temperature. Based on the results of the study, Multiple regression models were developed to estimate the mechanical properties of three-layer particleboard. The results of this study showed the mechanical properties of particleboard were highly related with resin content. specific gravity and mat moisture content in decending order. The mechanical properties were able to estimated as the linear function of resin content and specific gravity. However, the effects of change in mat moisture content on the mechanical properties showed a non-linear pattern. The mechanical properties curves over mat moisture content reached peaks at 15 %, and then decreased at 18 % and 21 % of mat moisture contents. On the other hand, the effects of pressing time and pressing temperature on the mechanical properties of particleboard were not significant.

• Korean Journal of Ecology and Environment
• /
• v.42 no.4
• /
• pp.432-440
• /
• 2009

During the period of past fifteen years (1992~2006), variations of chlorophyll a in relation with water quality in freshwater reservoirs were investigated. This study compared total nitrogen (TN), total phosphorus (TP), chlorophyll a, Secchi depth (SD) and total suspended solids (TSS) between terrestrial freshwater reservoir and coastal freshwater reservoir systems based on their location. Regression analyses (linear and non-linear regressions) were applied for all study sites to examine relationship and interaction of these factors in the freshwater systems from in-land to coasts. The results demonstrated that chlorophyll a was significantly correlated to total phosphorus ($R^2=0.94$, P<0.0001) and was remarkably related to TSS increase ($R^2=0.63$, P<0.0001) in the selected reservoirs. The TN : TP ratio in the reservoir systems was higher than Redfield ratio (16 : 1) indicating that the reservoirs are potentially experiencing P limitation. Water quality of coastal freshwater reservoir system was more significantly decreased than the reservoirs located in in-land during the past fifteen years. The strict management of nutrient discharge into freshwater systems should implemented in the coastal reservoirs since the freshwater is introduced into coastal estuarine systems.

• Bulletin of the Korean Chemical Society
• /
• v.29 no.4
• /
• pp.833-841
• /
• 2008

Principal component-genetic algorithm-multiparameter linear regression (PC-GA-MLR) and principal component-genetic algorithm-artificial neural network (PC-GA-ANN) models were applied for prediction of melting point for 323 drug-like compounds. A large number of theoretical descriptors were calculated for each compound. The first 234 principal components (PC’s) were found to explain more than 99.9% of variances in the original data matrix. From the pool of these PC’s, the genetic algorithm was employed for selection of the best set of extracted PC’s for PC-MLR and PC-ANN models. The models were generated using fifteen PC’s as variables. For evaluation of the predictive power of the models, melting points of 64 compounds in the prediction set were calculated. Root-mean square errors (RMSE) for PC-GA-MLR and PC-GA-ANN models are 48.18 and $12.77{^{\circ}C}$, respectively. Comparison of the results obtained by the models reveals superiority of the PC-GA-ANN relative to the PC-GA-MLR and the recently proposed models (RMSE = $40.7{^{\circ}C}$). The improvements are due to the fact that the melting point of the compounds demonstrates non-linear correlations with the principal components.

• Bulletin of the Korean Chemical Society
• /
• v.28 no.9
• /
• pp.1472-1476
• /
• 2007

Artificial neural networks (ANNs) are successfully developed for the modeling and prediction of normalized polarity parameter (ETN) of 216 various solvents with diverse chemical structures using a quantitative-structure property relationship. ANN with architecture 5-9-1 is generated using five molecular descriptors appearing in the multi-parameter linear regression (MLR) model. The most positive charge of a hydrogen atom (q+), total charge in molecule (qt), molecular volume of solvent (Vm), dipole moment (μ) and polarizability term (πI) are input descriptors and its output is ETN. It is found that properly selected and trained neural network with 192 solvents could fairly represent the dependence of normalized polarity parameter on molecular descriptors. For evaluation of the predictive power of the generated ANN, an optimized network is applied for prediction of the ETN values of 24 solvents in the prediction set, which are not used in the optimization procedure. Correlation coefficient (R) and root mean square error (RMSE) of 0.903 and 0.0887 for prediction set by MLR model should be compared with the values of 0.985 and 0.0375 by ANN model. These improvements are due to the fact that the ETN of solvents shows non-linear correlations with the molecular descriptors.

• Journal of the Korean Chemical Society
• /
• v.60 no.4
• /
• pp.225-234
• /
• 2016

The activity of 34 sulfonamide derivatives has been estimated by means of multiple linear regression (MLR), artificial neural network (ANN), simulated annealing (SA) and genetic algorithm (GA) techniques. These models were also utilized to select the most efficient subsets of descriptors in a cross-validation procedure for non-linear -log (IC50) prediction. The results obtained using GA-ANN were compared with MLR-MLR, MLR-ANN, SA-ANN and GA-ANN approaches. A high predictive ability was observed for the MLR-MLR, MLR-ANN, SA-ANN and MLR-GA models, with root mean sum square errors (RMSE) of 0.3958, 0.1006, 0.0359, 0.0326 and 0.0282 in gas phase and 0.2871, 0.0475, 0.0268, 0.0376 and 0.0097 in solvent, respectively (N=34). The results obtained using the GA-ANN method indicated that the activity of derivatives of sulfonamides depends on different parameters including DP03, BID, AAC, RDF035v, JGI9, TIE, R7e+, BELM6 descriptors in gas phase and Mor 32u, ESpm03d, RDF070v, ATS8m, MATS2e and R4p, L1u and R3m in solvent. In conclusion, the comparison of the quality of the ANN with different MLR models showed that ANN has a better predictive ability.

• Journal of Korea Planning Association
• /
• v.54 no.3
• /
• pp.15-26
• /
• 2019

The development of science and transportation technology leads to the increase of inter - city networks that play an important role in urban growth. Overall, numerous studies based on network theory pay attention to positive effects of urban network on urban growth. However, some studies have pointed out the negative effects of inter-city interactions such as straw effects. This implies that the network between cities may not be positively correlated with urban growth, and that the direction of the influence may vary from a certain threshold, such as the marginal utility curve. In this context, the purpose of this study is to measure the impacts of network with central city on urban growth in the capital region and examine the relationship between urban network and growth. Two multiple regression models are employed with changes in population and employment as dependent variables. The urban network index and other control variables are used as independent variables. Especially, the urban network indexes are used in quadratic forms to examine non linear relations with urban growth such U-shape or an inverted U-shape. The results show that the relationships between networks with the central city and urban growth are not a simple linear, and the influence can be changed from the critical point.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.4
• /
• pp.1209-1223
• /
• 2022

Blood pressure is one of the key physiological parameters for determining human health, and can prove whether human cardiovascular function is healthy or not. In general, what we call blood pressure refers to arterial blood pressure. Blood pressure fluctuates greatly and, due to the influence of various factors, even varies with each heartbeat. Therefore, achievement of continuous blood pressure measurement is particularly important for more accurate diagnosis. It is difficult to achieve long-term continuous blood pressure monitoring with traditional measurement methods due to the continuous wear of measuring instruments. On the other hand, radar technology is not easily affected by environmental factors and is capable of strong penetration. In this study, by using machine learning, tried to develop a linear blood pressure prediction model using data from a public database. The radar sensor evaluates the measured object, obtains the pulse waveform data, calculates the pulse transmission time, and obtains the blood pressure data through linear model regression analysis. Confirm its availability to facilitate follow-up research, such as integrating other sensors, collecting temperature, heartbeat, respiratory pulse and other data, and seeking medical treatment in time in case of abnormalities.

• Steel and Composite Structures
• /
• v.43 no.1
• /
• pp.107-116
• /
• 2022

This paper presents results of numerical studies completed on unreinforced and doubler plate reinforced Elliptical Hollow Section (EHS) T-joints subjected to axial compressive loading on the brace member. Non-linear finite element (FE) models were developed using the finite element code, ABAQUS. Available test data in literature was used to validate the FE models. Subsequently, a parametric study was carried out to investigate the effects of various geometrical parameters of main members and reinforcement plates on the ultimate capacity of reinforced EHS T-joints. The parametric study found that the reinforcing plate significantly increases the ultimate capacity of EHS T-joints up to twice the capacity of the corresponding unreinforced joint. The thickness and length of the reinforcing plate have a positive effect on the ultimate capacity of Type 1 joints. This study, however, found that the capacity of Type 1 orientation is not dependent on the brace-to-chord diameter ratio. As for type 2 orientations, the thickness and length of the reinforcement have a minimal effect on the ultimate capacity. A new design method is introduced to predict the capacity of the reinforced EHS T-joints Type 1 and 2 based on the multiple linear regression analyses.

• Korea Journal of Hospital Management
• /
• v.28 no.1
• /
• pp.37-51
• /
• 2023

Purposes: There exist many non-covered services that the National Health Insurance does not cover, and thus, their prices are set by individual health care providers. However, little study has been done to investigate how hospitals set prices for those services. The purpose of this study is to examine the relationship between ownership, profitability, and prices of those services for a sample of general hospitals. Methodology/Approach: Data regarding the prices of major non-covered services (e.g., upper-level hospital room fees, MRI, Da 7inci robot surgery, and LASIK) were obtained from the Health Insurance Review and Assessment Service and the financial information, as well as other characteristics, were derived from the financial reports from the Korea Health Industry Development Institute. Descriptive statistics, t-tests, and multiple linear regression analyses were used to test the relationship between the independent variables and the dependent variables. Findings: Hospitals owned by private universities appeared to have higher prices for non-covered services while regional public hospitals tend to have lower prices. Profitability, measured by operating margin, was not significantly related to the prices. Hospitals that charge higher prices were more likely to be located in the capital area (Seoul, Incheon, and Gyeonggi), and to employ larger number of personnel. Practical Implications: Public hospitals tend to charge lower prices for non-covered services. Relative market power appears to be related to pricing. Further research is needed to investigate whether such a relationship varies over time and its effects on the quality and access.