• Asia pacific journal of information systems
• /
• v.21 no.2
• /
• pp.43-58
• /
• 2011

Support vector machines (SVMs), a machine learning technique, has been applied to not only binary classification problems such as bankruptcy prediction but also multi-class problems such as corporate credit ratings. However, in general, the performance of SVMs can be easily worse than the best alternative model to SVMs according to the selection of predictors, even though SVMs has the distinguishing feature of successfully classifying and predicting in a lot of dichotomous or multi-class problems. For overcoming the weakness of SVMs, this study has proposed an approach for selecting features for multi-class SVMs that utilize the impurity measures of classification trees. For the selection of the input features, we employed the C4.5 and CART algorithms, including the stepwise method of discriminant analysis, which is a well-known method for selecting features. We have built a multi-class SVMs model for credit rating using the above method and presented experimental results with data regarding S&P 500 companies.

• Communications for Statistical Applications and Methods
• /
• v.2 no.1
• /
• pp.74-88
• /
• 1995

This article introduces a unified method of choosing the most explanatory and significant multiway partitions for classification tree design and analysis. The method is derived on the impurity reduction (IR) measure of divergence, which is proposed to extend the proportional-reduction-in-error (PRE) measure in the decision-theory context. For the method derivation, the IR measure is analyzed to characterize its statistical properties which are used to consistently handle the subjects of feature formation, feature selection, and feature deletion required in the associated classification tree construction. A numerical example is considered to illustrate the proposed approach.

• YAKHAK HOEJI
• /
• v.37 no.1
• /
• pp.76-83
• /
• 1993

The changes in discharge current, emission and/or oscillation frequency of the electric oscillation of a glow discharge are the potential sensitive measure of the concentration of an impurity in the argon plasma supporting gas. A single jet enhanced glow discharge has been interfaced with the gas chromatograph via 1/8" O.D. tube with a heating pad to study the changes in discharge current. To investigate the optimum operating conditions of the glow discharge system as detector for gas chromatography, pressure, gas flow rate, discharge current, distance between the anode and the cathode have been studied.

• Communications for Statistical Applications and Methods
• /
• v.11 no.3
• /
• pp.553-562
• /
• 2004

In this short communication we discuss the bias problems of CART in split variable selection and suggest a method to reduce the variable selection bias. Penalties proportional to the number of categories or distinct values are applied to the splitting criteria of CART. The results of empirical comparisons show that the proposed modification of CART reduces the bias in variable selection.

• Mass Spectrometry Letters
• /
• v.1 no.1
• /
• pp.17-20
• /
• 2010

As background significantly affects measurement accuracy and a detection limit in determination of the trace amounts of uranium, it is necessary to minimize the impurities in the filaments used for thermal ionization mass spectrometry (TIMS). We have varied the degassing condition such as the heating currents and duration times to reduce the backgrounds from the filaments prepared with zone-refined rhenium tape. The most efficient degassing condition of the heating current and the duration time was determined as 3.5 A and 60 min, respectively. The TIMS measurement combined with the isotope dilution mass spectrometry (IDMS) technique showed that the uranium backgrounds were determined to be in a few fg level from blank rhenium filaments. The background minimized filaments were utilized to measure the uranium isotope ratios of a U030 (NIST) standard sample. The excellent agreement of the measurement with the certified isotope ratios showed that the degassing procedure optimized in this study efficiently reduced the impurity signals of uranium from blank rhenium filaments to a negligible level.

• Proceedings of the Korean Society For Composite Materials Conference
• /
• 2005.04a
• /
• pp.265-268
• /
• 2005

A new method for measurement of the pore size in a nanohoneycomb structure using atomic force microscopy (AFM) was proposed. Porous type anodic aluminum oxide (AAO) was fabricated as a nanohoneycomb structure to measure the pore size. For measuring pore sizes from AFM images, a criterion was set in porous type AAO. The pore sizes from AFM images were compared with those from SEM images, and the results showed good agreement. The relationship between the pore size and widening time was found to be linear in the range of this study. It was understood as the synchronized effects of the impurity gradient in outer oxide of AAO, mechanical packing and mass transfer increase.

• Proceedings of the KIEE Conference
• /
• 2008.09a
• /
• pp.235-236
• /
• 2008

Charging of spacecraft occurs in plasma and radiation environment. Especially, we focused on an accident caused by internal charging in a glass material that was used as the cover plate of solar panel array, and tried to measure the charge distribution in glass materials under electron beam irradiation by using a PEA (Pulsed Electro-Acoustic method) system. In the case of a quartz glass (pure $SiO_2$), no charge accumulation was observed either during or after the electron beam irradiation. On the contrary, positive charge accumulation was observed in glass samples containing metal-oxide components. It is found that the polarity of the observed charges depends on the contents of the impurities. To identify which impurity dominates the polarity of the accumulated charge, we measured charge distributions in several glass materials containing various metal-oxide components and calculated the trap energy depths from the charge decay characteristics of all glass samples.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.18 no.1
• /
• pp.7-19
• /
• 1981

Anodization method is utilized in order to measure diffusion profiles of boron and arsenic in silicon. The solution used for silicon anodization is Ethylene glycol +KNO3(0.04N), The thickness of silicon which is consumed by a single 200V anodization is 460$\pm$40A regardless of wafer type. The profiles of boron and arsenic in silicon after predeposition process are investigated. The diffusion coefficients of both dopants depending on impurity concentration are extrated from these profiles. The base pull-in effect has been observed in prototype npn transistors with arsenic doped emitter.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.2305-2306
• /
• 2008

Charging of spacecraft occurs in plasma and radiation environment. Especially, we focused on an accident caused by internal charging in a glass material that was used as the cover plate of solar panel array, and tried to measure the charge distribution in glass materials under electron beam irradiation by using a PEA (Pulsed Electro-Acoustic method) system. In the case of a quartz glass (pure $SiO_2$), no charge accumulation was observed either during or after the electron beam irradiation. On the contrary, positive charge accumulation was observed in glass samples containing metal-oxide components. It is found that the polarity of the observed charges depends on the contents of the impurities. To identify which impurity dominates the polarity of the accumulated charge, we measured charge distributions in several glass materials containing various metal-oxide components and calculated the trap energy depths from the charge decay characteristics of all glass samples.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2008.11a
• /
• pp.268-269
• /
• 2008

Charging of spacecraft occurs in plasma and radiation environment. Especially, we focused on an accident caused by internal charging in a glass material that was used as the cover plate of solar panel array, and tried to measure the charge distribution in glass materials under electron beam irradiation by using a PEA (Pulsed Electro-Acoustic method) system. In the case of a quartz glass (pure $SiO_2$), no charge accumulation was observed either during or after the electron beam irradiation. On the contrary, positive charge accumulation was observed in glass samples containing metal-oxide components. It is found that the polarity of the observed charges depends on the contents of the impurities. To identify which impurity dominates the polarity of the accumulated charge, we measured charge distributions in several glass materials containing various metal-oxide components and calculated the trap energy depths from the charge decay characteristics of all glass samples.