• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.3
• /
• pp.77-90
• /
• 2018

Software Defect Prediction (SDP) is a field of study that identifies defective modules. With insufficient local data, a company can exploit Cross-Project Defect Prediction (CPDP), a way to build a classifier using dataset collected from other companies. Most machine learning algorithms for SDP have used more than one parameter that significantly affects prediction performance depending on different values. The objective of this study is to propose a parameter selection technique to enhance the performance of CPDP. Using a Harmony Search algorithm (HS), our approach tunes parameters of cost-sensitive boosting, a method to tackle class imbalance causing the difficulty of prediction. According to distributional characteristics, parameter ranges and constraint rules between parameters are defined and applied to HS. The proposed approach is compared with three CPDP methods and a Within-Project Defect Prediction (WPDP) method over fifteen target projects. The experimental results indicate that the proposed model outperforms the other CPDP methods in the context of class imbalance. Unlike the previous researches showing high probability of false alarm or low probability of detection, our approach provides acceptable high PD and low PF while providing high overall performance. It also provides similar performance compared with WPDP.

• Journal of Korean Society of Environmental Engineers
• /
• v.30 no.6
• /
• pp.621-627
• /
• 2008

In this study, a technique of speciation and determination of the trace inorganic arsenic(As(III) and As(V)) in water sample using HPLC-DRC-ICP-MS has been developed. Isocratic mobile phase of 10 mM ammonium nitrate and 10 mM ammonium phosphate monobasic was used and methanol(5 v/v%) was used as flushing solvent. Selection of the best flow rate of reaction gas, O$_2$, and optimization of the parameters such as pH and flow rate of mobile phase, and injection volume of sample for the separation and detection of arsenic species were carried out. The oxygen flow rate of 0.5 mL/min, pH of 9.4 and flow rate of 1.5 mL/min of mobile phase, and injection volume of sample of 100 $\mu$L were found to be the best parameters for the speciation and determination of arsenic species. The analytical features of the method were detection limit 0.10 and 0.08 $\mu$g/L, precision(RSD) 4.3% and 3.6%, and recovery 95.2% and 96.4% for As(III) and As(V), respectively. Analysis time was 4 minutes per sample. Linear calibration graphs with r$^2$ = 0.998 were obtained for both As(III) and As(V). Speciation analysis of arsenic species in the raw water samples collected from the tributary streams to Han River and main stream of Paldnag were performed by the proposed method. The concentrations of As(III) ranged from 0.10 to 0.22 $\mu$g/L and As(V) concentrations ranged from 0.44 to 1.19 $\mu$g/L, and 93.5% of total arsenic was found to be As(V).

• Journal of the Korean Applied Science and Technology
• /
• v.34 no.3
• /
• pp.451-460
• /
• 2017

This study optimized the extraction of antioxidants from Gugija (Lycium chinensis Mill). To determine operational parameters, including ethanol concentration ($X_1$, 0~80%) and extraction time ($X_2$, 1~5 hr), response surface methodology was applied to monitor yield, anthocyanins, flavonoids and DPPH radical scavenging activity. Coefficients of determinations ($R^2$) of the models were range of 0.8645~0.9859 (p<0.01~0.1) in dependant parameters. Yield of Gugija extracts was maximized 23.12% in extraction conditions of 4.22 h at 8.25% ethanol. Anthocyanins was maximized 1.43 (OD in 530 nm) in extraction conditions of 3.06 h at 79.98% ethanol. Flavonoids was maximized $3,100{\mu}g/100g$ in extraction conditions of 3.37 h at 67.02% ethanol. DPPH radical scavenging activity was maximized 96.93% in extraction conditions of 1.67 h at 69.81% ethanol. Optimum extraction conditions (2.5 h extraction at 70% ethanol) were obtained by superimposing the contour maps with regard to anthocyanins, flavonoids and DPPH radical scavenging activity of Gugija. Maximum values of anthocyanins, flavonoids and DPPH radical scavenging activity in optimum extraction condition were 1.0080 (OD in 530 nm), $3,145{\mu}g/100g$, 96.96%, respectively. But values of anthocyanins, flavonoids and DPPH radical scavenging activity in water extraction condition (1 h at water) were 0.4652 (OD in 530 nm), $1,633{\mu}g/100g$, 86.98%, respectively.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.11
• /
• pp.483-490
• /
• 2016

Automobile door latch is a fine design and assembly techniques are required in order to produce them in a small component assembly shape such as a spring, injection products, a small-sized motor. The door latch is fixed to not open the door of the car plays an important role it has a direct impact on the driver's safety. In this study, during trimming of the terminals of the connector main components of the car door latch, reduce rollover and conducted a research to find a suitable effective shear surface. Using the Taguchi method with orthogonal array of Finite Element Analysis and optimal Design of Experiments were set up parameters for the shear surface quality of the car door latch connector terminals. The design parameters used in the analysis is the clearance, the radius, and the blank holding force, the material of the connector terminal is a C2600. Trimming process optimum conditions suggested by the analysis has been verified by experiments, the shear surface shape and dimensions of a final product in good agreement with forming analysis results.Taguchi method from the above results in the optimization for the final rollover and effective shear surface improved for a vehicle door latch to the connector terminal can be seen that the applicable and useful for a variety of metal forming processes other than the trimming process is determined to be applicable.

• Journal of Korea Water Resources Association
• /
• v.44 no.9
• /
• pp.711-720
• /
• 2011

SWAT (Soil and Water Assessment Tool) model was calibrated for the flow rate of the Deachung lake with a large area of 3108.29 $km^2$. Application of SWAT model requires significant number of input data and is prone to result in uncertainties due to errors in input data, model structure and model parameters. The SUFI-2 (Sequential Uncertainty Fitting Ver. 2) program and GLUE (Generalized Likelihood Uncertainty Estimation) program in SWAT-CUP (SWAT-Calibration and Uncertainty Program) are used to select the best parameters for SWAT model. Optimal combination of parameter values was determined through 2,000 iterative SWAT model runs. The Nash-Sutcliffe values and $R^2$ values were 0.87 and 0.89 respectively indicating both methods show good agreements with observed data successfully. RMSE and MSE values also showed similar results for both programs. It seems the SWAT-CUP has a great practical appeal for parameter optimization especially for large basin area and it also can be used for less experienced SWAT model users.

• Journal of Korea Water Resources Association
• /
• v.43 no.10
• /
• pp.893-909
• /
• 2010

The purpose of this study is to consider the interaction between surface water and groundwater in basin scale by developing TANK_GS model. The soil moisture structure of tank model with 3 tanks is improved to simulate the appropriate stream-aquifer interactions. Maximum likelihood method is applied to calibrate parameters with variance functions to deal with heteroscedasticity of residuals. The parameters of improved TANK_GS model and variance function are simultaneously estimated by Simulated Annealing method, a global optimization technique. The results of TANK-GE are compared to those of the SWMM-GE model which had been developed to consider the stream-aquifer interactions. The new TANK_GS model and SWMM-GE model are applied to Gapcheon basin, which belongs to Geum River basin. TANK_GS model showed better model performance compared to the original TANK model and characterized the relationship of stream-aquifer interactions as satisfactorily as the SWMM-GE model. The sustainable groundwater yield can be estimated for the regional water resources planning using the TANK_GS model

• Journal of Korea Water Resources Association
• /
• v.47 no.4
• /
• pp.343-356
• /
• 2014

The SWMM (storm water management model) has been widely used in the world and is a watershed runoff simulation model used for a single event or a continuous simulation of runoff quantity and quality. However, there are many uncertain parameters in the watershed runoff continuous simulation module and the water quality module, which make it difficult to use the SWMM. The purpose of the study is to develop an automatic calibration module of the SWMM not only for watershed runoff continuous simulation, but also water quality simulation. The automatic calibration module was developed by linking the SWMM with the SCE-UA (shuffled complex evolution-University of Arizona) that is a global optimization algorithm. Estimation parameters of the SWMM were selected and search ranges of them were reasonably configured. The module was validated by calibration and verification of the watershed runoff continuous simulation model and the water quality model for the Donghyang Stage Station Basin. The calibration results for watershed runoff continuous simulation model were excellent and those for water quality simulation model were generally satisfactory. The module could be used in various studies and designs for watershed runoff and water quality analyses.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.34 no.4
• /
• pp.1105-1116
• /
• 2014

A stochastic rainfall model, NSRPM (Neyman-Scott Rectangular Pulse Model), is able to reflect the cluster characteristics of rainfall events which is unable in the RPM (Rectangular Pulse Model). Therefore NSRPM has advantage in the hydrological applications. The NSRPM consists of five model parameters and the parameters are estimated using optimization techniques such as DFP (Davidon-Fletcher-Powell) method and genetic algorithm. However the DFP method is very sensitive in initial values and is easily converge to local minimum. Also genetic algorithm has disadvantage of long computation time. Nelder-Mead method has several advantages of short computation time and no need of a proper initial value. In this study, the applicability of parameter estimation methods was evaluated using rainfall data of 59 national rainfall networks from 1973-2011. Overall results demonstrated that accuracy in parameter estimation is in the order of Nelder-Mead method, genetic algorithm, and DFP method.

• Journal of the Korean Chemical Society
• /
• v.54 no.3
• /
• pp.269-274
• /
• 2010

The solubility and the physicochemical property (refractive index) in the NaCl-$CaCl_2$-$H_2O$ and KCl-$CaCl_2$-$H_2O$ systems were determined at $50^{\circ}C$ and the phase diagrams and the diagrams of physicochemical property vs composition were plotted. One invariant point, two univariant curves, and two crystallization zones, corresponding to sodium Chloride (or potassium chloride), dihydrate ($CaCl_2{\cdot}2H_2O$) showed up in the phase diagrams of the ternary systems. The mixing parameters ${\theta}_{M,Ca}$ and ${\Psi}_{M,Ca,Cl}$ (M = Na or K) and equilibrium constant $K_{sp}$ were evaluated in NaCl-$CaCl_2-H_2O$ and KCl-$CaCl_2-H_2O$ systems by least-squares optimization procedure, in which the single-salt Pitzer parameters of NaCl, KCl and $CaCl_2$ ${\beta}^{(0)}$, ${\beta}^{(1)}$, ${\beta}^{(2)}$ and $C^{\Phi}$ were directly calculated from the literature. The results obtained were in good agreement with the experimental data.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.5
• /
• pp.487-496
• /
• 2000

In this paper, the Multi-FNN(Fuzzy-Neural Networks) model is identified and optimized using HCM(Hard C-Means) clustering method and genetic algorithms. The proposed Multi-FNN is based on Yamakawa's FNN and uses simplified inference as fuzzy inference method and error back propagation algorithm as learning rules. We use a HCM clustering and Genetic Algorithms(GAs) to identify both the structure and the parameters of a Multi-FNN model. Here, HCM clustering method, which is carried out for the process data preprocessing of system modeling, is utilized to determine the structure of Multi-FNN according to the divisions of input-output space using I/O process data. Also, the parameters of Multi-FNN model such as apexes of membership function, learning rates and momentum coefficients are adjusted using genetic algorithms. A aggregate performance index with a weighting factor is used to achieve a sound balance between approximation and generalization abilities of the model. The aggregate performance index stands for an aggregate objective function with a weighting factor to consider a mutual balance and dependency between approximation and predictive abilities. According to the selection and adjustment of a weighting factor of this aggregate abjective function which depends on the number of data and a certain degree of nonlinearity, we show that it is available and effective to design an optimal Multi-FNN model. To evaluate the performance of the proposed model, we use the time series data for gas furnace and the numerical data of nonlinear function.