• Journal of Information Processing Systems
• /
• v.10 no.2
• /
• pp.314-333
• /
• 2014

When analyzing default predictions in real estate companies, the number of non-defaulted cases always greatly exceeds the defaulted ones, which creates the two-class imbalance problem. This lowers the ability of prediction models to distinguish the default sample. In order to avoid this sample selection bias and to improve the prediction model, this paper applies a minority sample generation approach to create new minority samples. The logistic regression, support vector machine (SVM) classification, and neural network (NN) classification use an imbalanced dataset. They were used as benchmarks with a single prediction model that used a balanced dataset corrected by the minority samples generation approach. Instead of using prediction-oriented tests and the overall accuracy, the true positive rate (TPR), the true negative rate (TNR), G-mean, and F-score are used to measure the performance of default prediction models for imbalanced dataset. In this paper, we describe an empirical experiment that used a sampling of 14 default and 315 non-default listed real estate companies in China and report that most results using single prediction models with a balanced dataset generated better results than an imbalanced dataset.

• Analytical Science and Technology
• /
• v.20 no.1
• /
• pp.55-60
• /
• 2007

A sensitive technique for the determination of trace Cu(II) in various samples after the column preconcentration by adsorbing its N-benzoylphenylhydroxylamine (BPHA) onto benzophenone was developed. Several experimental conditions such as the pH of the sample solution, the amount of chelating agent, the amount of benzophenone, and the flowrate of sample solution and so forth were optimized. The interfering effects of diverse concomitant ions were investigated. Fe(III) and $CN^-$ interfered with more seriously than any other ions. However, the interference by these ions could be overcome sufficiently by adjusting the added volume of 0.01M BPHA to 10 mL. The dynamic range, the correlation coefficient ($r^2$) and the detection limit obtained by this proposed technique were 5.0~120 ng/mL, 0.9974, and 2.1 ng/mL, respectively. For validating this proposed technique, the aqueous samples (stream water, reservoir water, and wastewater), the plastic sample and the diluted brass sample were used. Recovery yields of 93~102% were obtained. These measured data were not different from ICP-MS data at 95% confidence level. This method was also validated by the rice flour CRM (normal, fortified) samples. Based on the results from the experiment, it was found that this proposed technique could be applied to the determination of Cu(II) in various real samples.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.43 no.6
• /
• pp.135-142
• /
• 2001

Laboratory experiment was performed to evaluate the influence of mixing conditions to the strength of solidified sandy soils with cement. The major physical factors considered in this experiment were the fine particles content(<$\sharp200%$), cement content(%) and water-cement ratio, and unconfined compressive strength test was performed on the samples at 7 and 28 cured day. The results of tests shows that when the cement content is relatively low (7~10 percents) the fine content in the sandy soils is very important, but when cement content is high the water-cement ratio became more important. It was appeared that in the range of the cement content of 7~10 percents, about 20~30 percents of fine content to the total sample weight is the optimum condition to get the maximum strength. In the case of the cement content of 13 percents, the strength of sample was considerably affected by the water-cement ratio rather than the fine content. In this paper, empirical equations were also developed and evaluated to verify the relationship among three factors by the multi-regression analysis.

• Textile Coloration and Finishing
• /
• v.18 no.2 s.87
• /
• pp.15-23
• /
• 2006

A new fastness formula based on CIEDE2000 color-difference formula is developed by B. Rigg and his coworkers. It is very simple to calculate fastness grade for color change than ISO 105-A05 fastness formula based on CIELAB color-difference formula. Sample pair sets which cover a wide range color space were accumulated from NCS(Natural Color System) color book. For those sample pair sets, visual measurement experiment and instrument measurement experiment of fastness grade were carried out and each performance of ISO 105-A02 fastness formula and newly developed fastness formula was compared through degree of agreement for visual measurement result. Newly developed fastness formula indicated improved performance for measuring fastness grade but current ISO fastness formula for assessing change in color, ISO 105-A05, was confirmed that it's performance is inadequate to measure fastness grade. Then fastness formulae were examined more closely according to particular color spaces and the correlation of hue, lightness and chrom for measuring fastness grade was also considered in this study.

• Journal of the Korean Society for information Management
• /
• v.8 no.1
• /
• pp.100-116
• /
• 1991

In this study, a retrieval system taking advantage of term relationship in a specific domain and also of evidential reasoning as tools for measuring relevance is implemented. For this experiment, techincal memoranda documented in Electronics and Telecommunications Research Institute (ETRI) served as a sample document file. Sample knowledge base was prepared by extracting terms and term relations pertaining to telecommunications from INSPEC thesaurus. Relations between terms were represented by numerical values according to types of term relations. Relationship between a query and a document was measured according to Dempster-Shafer theory of evidence. As a result of this experiment, a more comprehensive search was made by expanding search terms utilizing term relations. Measure of relevance represented by reflecting term relations, and search results were listed in a descending order of relevance.

• Journal of Sensor Science and Technology
• /
• v.25 no.6
• /
• pp.431-434
• /
• 2016

Currently, the radial basis function network (RBFN) and various other neural networks are employed to classify gases using chemical sensors arrays, and their performance is steadily improving. In particular, the identification performance of the RBFN algorithm is being improved by optimizing parameters such as the center, width, and weight, and improved algorithms such as the radial basis function network-stochastic gradient (RBFN-SG) and radial basis function network-normalized stochastic gradient (RBFN-NSG) have been announced. In this study, we optimized the number of centers, which is one of the parameters of the RBFN-NSG algorithm, and observed the change in the identification performance. For the experiment, repeated measurement data of 8 samples were used, and the elbow method was applied to determine the optimal number of centers for each sample of input data. The experiment was carried out in two cases(the only one center per sample and the optimal number of centers obtained by elbow method), and the experimental results were compared using the mean square error (MSE). From the results of the experiments, we observed that the case having an optimal number of centers, obtained using the elbow method, showed a better identification performance than that without any optimization.

• Resources Recycling
• /
• v.11 no.3
• /
• pp.37-45
• /
• 2002

Triboelectrostatic Separation procese is a technology that different particles charged after contact and rubbing different materials are separated in an electric field. At this time, charged polarity of different materials depends on their own work function. Therefore this study discovers work functions of various plastics and determines charging characteristics for the specific optimum operation condition. The experiment is conducted with two sample sets composed of various different plastics. Each sample is charged by contact and rubbing different materials. Surface charge of charged particles is measured by Faraday Cage. The specific work function of an each plastic is driven by measured charging amount and charged particles are separated in a certain electric field ($\pm$20 kV). At last, the relationship between charging amount and separation efficiency is induced by the separation experiment.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.21 no.7
• /
• pp.386-393
• /
• 2009

The present study has been studied on a thermal and flow characteristic of the microchannel waterblock with flow distributions in each channels. Results of a numerical analysis using the CFX-11 are compared with results of an experiment. Numerical analysis and experiment are conducted under an input power of 150 W, inlet temperature of $20^{\circ}C$ and mass flow rates of $0.7{\sim}2.0$ kg/min. Base temperature and pressure drop are investigated with standard deviations of mass flow rates in each channels of samples. The flow distribution and j/f factor of the sample 4 is increased by about 65.7% and 42.6%, compared to that of the reference model sample 3.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.6 no.2
• /
• pp.803-808
• /
• 1964

In this experiment, it was attempted to determine the thermal conductivities of earth block materials which are recently much available for farm building constructions in Korea as lower cost building materials. For this experiment, two kinds of earth block materials were sampled. One of them was pure earth and the other was cementstabilized earth with a ratio of earth to cement by volume, 20 to 1. Cylindrical sections of specimen surrounding a steel pipe(I.D. =1 inch) were used. A resistance neater in the pipe and thermometers for the measurement of radial temperature distribution were installed as shown in Fig. 1 and photograph 1. The heat which flows through the pipe and the sample was produced by passing a current through a resistance wire stretched along the pipe axis. The complete apparatus used in this experiment is schematically shown in Fig. 1.

• Journal of Korea Foresty Energy
• /
• v.21 no.3
• /
• pp.54-58
• /
• 2002

White birch bark is being called baekhwapi usually and is used urination, travail, removal of fever in Oriental medicine. Betulin had been known as main chemistry ingredient of white birch bark. So, in an experiment that see betulin and connection inorganic compounds separation and enforced to recognize antioxidative activity and anti HIV. Betulin's yield separated vast quantity for the first time in Korea by 6.9%, and 3,4-dihydrobenzoic acid in ethylacetate soluble as experiment result. Wish to utilize betulin connection derivative as each kind experiment sample ore because do composition making since.