• Asia pacific journal of information systems
• /
• v.19 no.2
• /
• pp.157-178
• /
• 2009

Corporate credit rating is a very important factor in the market for corporate debt. Information concerning corporate operations is often disseminated to market participants through the changes in credit ratings that are published by professional rating agencies, such as Standard and Poor's (S&P) and Moody's Investor Service. Since these agencies generally require a large fee for the service, and the periodically provided ratings sometimes do not reflect the default risk of the company at the time, it may be advantageous for bond-market participants to be able to classify credit ratings before the agencies actually publish them. As a result, it is very important for companies (especially, financial companies) to develop a proper model of credit rating. From a technical perspective, the credit rating constitutes a typical, multiclass, classification problem because rating agencies generally have ten or more categories of ratings. For example, S&P's ratings range from AAA for the highest-quality bonds to D for the lowest-quality bonds. The professional rating agencies emphasize the importance of analysts' subjective judgments in the determination of credit ratings. However, in practice, a mathematical model that uses the financial variables of companies plays an important role in determining credit ratings, since it is convenient to apply and cost efficient. These financial variables include the ratios that represent a company's leverage status, liquidity status, and profitability status. Several statistical and artificial intelligence (AI) techniques have been applied as tools for predicting credit ratings. Among them, artificial neural networks are most prevalent in the area of finance because of their broad applicability to many business problems and their preeminent ability to adapt. However, artificial neural networks also have many defects, including the difficulty in determining the values of the control parameters and the number of processing elements in the layer as well as the risk of over-fitting. Of late, because of their robustness and high accuracy, support vector machines (SVMs) have become popular as a solution for problems with generating accurate prediction. An SVM's solution may be globally optimal because SVMs seek to minimize structural risk. On the other hand, artificial neural network models may tend to find locally optimal solutions because they seek to minimize empirical risk. In addition, no parameters need to be tuned in SVMs, barring the upper bound for non-separable cases in linear SVMs. Since SVMs were originally devised for binary classification, however they are not intrinsically geared for multiclass classifications as in credit ratings. Thus, researchers have tried to extend the original SVM to multiclass classification. Hitherto, a variety of techniques to extend standard SVMs to multiclass SVMs (MSVMs) has been proposed in the literature Only a few types of MSVM are, however, tested using prior studies that apply MSVMs to credit ratings studies. In this study, we examined six different techniques of MSVMs: (1) One-Against-One, (2) One-Against-AIL (3) DAGSVM, (4) ECOC, (5) Method of Weston and Watkins, and (6) Method of Crammer and Singer. In addition, we examined the prediction accuracy of some modified version of conventional MSVM techniques. To find the most appropriate technique of MSVMs for corporate bond rating, we applied all the techniques of MSVMs to a real-world case of credit rating in Korea. The best application is in corporate bond rating, which is the most frequently studied area of credit rating for specific debt issues or other financial obligations. For our study the research data were collected from National Information and Credit Evaluation, Inc., a major bond-rating company in Korea. The data set is comprised of the bond-ratings for the year 2002 and various financial variables for 1,295 companies from the manufacturing industry in Korea. We compared the results of these techniques with one another, and with those of traditional methods for credit ratings, such as multiple discriminant analysis (MDA), multinomial logistic regression (MLOGIT), and artificial neural networks (ANNs). As a result, we found that DAGSVM with an ordered list was the best approach for the prediction of bond rating. In addition, we found that the modified version of ECOC approach can yield higher prediction accuracy for the cases showing clear patterns.

• Journal of Korean Society for Atmospheric Environment
• /
• v.13 no.2
• /
• pp.147-159
• /
• 1997

One of the most critical problems associated with filter-pack data comparisons from various field networks is the use of different sampling flow rates. In this field study, the effects of various filter-pack(FP) sampling conditions were examined. Experiments were conducted to evaluate the effects of varying sampling flow rates (1.5 to 10 sLpm) on measured concentrations of dry acid deposition species. Collocated FP samples were also collected to determine sampling and analysis data reproducibility. Ambient air samples were collected simultaneously for the seven day durations at varous flow rate. The chemical species measured were sulfur dioxide ($SO_2$), particulate sulfate(P-$SO_{4}^{2+}$), nitric acid ($HNO_3$), and particulate nitrate (P-$NO_{3}^{-}$). The results indicated that the collocated samples can be measured reproducibly and that sampling bias for the species measured is not significant. It was concluded that variations in the flow rates (1.5 to 10 sLpm) did not significantly affect the concentration of the species of interest. Although the results were not significantly different between different flow rates, artifacts were more likely to occur at high flow conditions (>5 sLpm) (e.g., via volatilization of particulate nitrates) than at low flow conditions(<5 sLpm).

• Journal of Soil and Groundwater Environment
• /
• v.11 no.5
• /
• pp.20-34
• /
• 2006

This study was undertaken to assess the anthropogenic impact on trace metal concentrations (Zn, Cu, Pb, Cr, Ni, and Cd) of roadside sediments (N = 70) from No.7 national road within the watershed of Hoidong Reservoir in Pusan City and to estimate the potential mobility of selected metals using sequential extraction. We generally found high concentrations of metals, especially Zn, Cu and Pb, affected by anthropogenic inputs. Compared to the trace metal concentrations of uncontaminated stream sediments, arithmetic mean concentrations of roadside sediments were about 7 times higher for Cu, 4 times higher for Zn, 3 times higher for Pb and Cr and, 2 times higher for Ni and As. Speciation data on the basis of sequential extraction indicate that most of the trace metals considered do not occur in significant quantities in the exchangeable fraction, except for Cd and Ni whose exchangeable fractions are appreciable (average 29.3 and 25.8%, respectively). Other metals such as Zn (51.4%) and Pb (45.2%) are preferentially bound to the reducible fraction, and therefore they can be potentially released by a pH decrease and/or redox change. Copper is mainly found in the organic fraction, while Cd is highest in the exchangeable fraction, and Cr and Ni in the residual fraction. Considering the proportion of metals bound to the exchangeable and carbonate fractions, the comparative mobility of metals probably decreases in the order of Cd>Ni>Pb>Zn>Cr>Cu. Although the total concentration data showed that Zn was typically present in potentially harmful concentration levels, the data on metal partitioning indicated that Cd, Ni and Pb pose the highest potential hazard for runoff water. As potential changes of redox state and pH may remobilize the metals bound to carbonates, amorphous oxides, and/or organic matter, and may release and flush them through drain networks into the watershed of Hoidong Reservoir, careful monitoring of environmental conditions appears to be very important.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.2
• /
• pp.416-424
• /
• 2017

In this study, we propose a design of digits recognition system based on RBFNNs through a comparative study of pre-processing algorithms in order to recognize digits in handwritten. Histogram of Oriented Gradient(HOG) is used to get the features of digits in the proposed digits recognition system. In the pre-processing part, a dimensional reduction is executed by using Principal Component Analysis(PCA) and (2D)2PCA which are widely adopted methods in order to minimize a loss of the information during the reduction process of feature space. Also, The architecture of radial basis function neural networks consists of three functional modules such as condition, conclusion, and inference part. In the condition part, the input space is partitioned with the use of fuzzy clustering realized by means of the Fuzzy C-Means algorithm. Also, it is used instead of gaussian function to consider the characteristic of input data. In the conclusion part, the connection weights are used as the extended type of polynomial expression such as constant, linear, quadratic and modified quadratic. By using MNIST handwritten digit benchmarking database, experimental results show the effectiveness and efficiency of proposed digit recognition system when compared with other studies.

• Membrane and Water Treatment
• /
• v.9 no.2
• /
• pp.115-121
• /
• 2018

Artificial neural network (ANN) simulation is used to predict the dynamic change of permeate flux during wheat starch industry wastewater microfiltration with and without static turbulence promoter. The experimental program spans range of a sedimentation times from 2 to 4 h, for feed flow rates 50 to 150 L/h, at transmembrane pressures covering the range of $1{\times}10^5$ to $3{\times}10^5Pa$. ANN predictions of the wastewater microfiltration are compared with experimental results obtained using two different set of microfiltration experiments, with and without static turbulence promoter. The effects of the training algorithm, neural network architectures on the ANN performance are discussed. For the most of the cases considered, the ANN proved to be an adequate interpolation tool, where an excellent prediction was obtained using automated Bayesian regularization as training algorithm. The optimal ANN architecture was determined as 4-10-1 with hyperbolic tangent sigmoid transfer function transfer function for hidden and output layers. The error distributions of data revealed that experimental results are in very good agreement with computed ones with only 2% data points had absolute relative error greater than 20% for the microfiltration without static turbulence promoter whereas for the microfiltration with static turbulence promoter it was 1%. The contribution of filtration time variable to flux values provided by ANNs was determined in an important level at the range of 52-66% due to increased membrane fouling by the time. In the case of microfiltration with static turbulence promoter, relative importance of transmembrane pressure and feed flow rate increased for about 30%.

• Journal of the Society of Disaster Information
• /
• v.1 no.1
• /
• pp.91-119
• /
• 2005

When the national disaster accident is occurred, it is difficult to maintain the mutual cooperation systems. In order to solve the problems, the construction of the national unified command wireless network is necessary. In this paper, the specified state of the characteristic frequency of the digital TRS wireless network constructed recently is investigated and analyzed. Through the analysis, the problems of the construction of the national unified command wireless network are grasped. To solve the problem, it is proposed that the digital TRS wireless network is connected with the satellite communication network, and connected with the existing wireless network, LMR. In the concretely it is proposed that the natural unified wireless network should be proceeded step by step. At first, for 2 years the existing networks of the Fire Fighting Agency, the Police, the Forest Service and so on must be utilized and prepared to link with TRS. The second, for 2 years it is carried forward a scheme to maintain the properties of the agencies concerned. Further, it must be prepared to connect with satellite network. At third, for 2 years all agencies concerned with the fire fighting and the disaster prevention must be unified, and the systems have to be promoted for the p1an of linkage of TRS network and the existing network. Next the agencies concerned have to be unified and the authority has to be intensified. When a disaster is occurred, the National Emergency Management Agency has to play a central role. In a local area it has to be given the Fire Fighting Agency an authority and a duty to get ready for each emergency situation.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.8 no.3
• /
• pp.91-99
• /
• 2009

Development of high-speed Internet services and the increased supply of mobile devices have become the key factor for the acceleration of ubiquitous technology. WiBro system, formed with lP backbone network, is a MBWA technology which provides high-speed multimedia service in a possibly broader coverage than Wireless LAN can offer. Wireless telecommunication environment needs not only mobility support in Layer 2 but also mobility management protocol in Layer 3 and has to minimize handover latency to provide seamless mobile services. In this paper, we propose a fast cross-layer handover scheme based on signal strength prediction in WiBro environment. The signal strength is measured at regular intervals and future value of the strength is predicted by Exponential Smoothing Method. With the help of the prediction, layer-3 handover activities are able to occur prior to layer-2 handover, and therefore, total handover latency is reduced. Simulation results demonstrate that the proposed scheme predicts that future signal level accurately and reduces the total handover latency.

• Molecules and Cells
• /
• v.42 no.6
• /
• pp.460-469
• /
• 2019

Bacterial ${\alpha}-type$ carbonic anhydrase (${\alpha}-CA$) is a zinc metalloenzyme that catalyzes the reversible and extremely rapid interconversion of carbon dioxide to bicarbonate. In this study, we report the first crystal structure of a hyperthermostable ${\alpha}-CA$ from Persephonella marina EX-H1 (pmCA) in the absence and presence of competitive inhibitor, acetazolamide. The structure reveals a compactly folded pmCA homodimer in which each monomer consists of a 10-stranded ${\beta}-sheet$ in the center. The catalytic zinc ion is coordinated by three highly conserved histidine residues with an exchangeable fourth ligand (a water molecule, a bicarbonate anion, or the sulfonamide group of acetazolamide). Together with an intramolecular disulfide bond, extensive interfacial networks of hydrogen bonds, ionic and hydrophobic interactions stabilize the dimeric structure and are likely responsible for the high thermal stability. We also identified novel binding sites for calcium ions at the crystallographic interface, which serve as molecular glue linking negatively charged and otherwise repulsive surfaces. Furthermore, this large negatively charged patch appears to further increase the thermostability at alkaline pH range via favorable charge-charge interactions between pmCA and solvent molecules. These findings may assist development of novel ${\alpha}-CAs$ with improved thermal and/or alkaline stability for applications such as $CO_2$ capture and sequestration.

• Journal of Biomedical Engineering Research
• /
• v.41 no.2
• /
• pp.107-119
• /
• 2020

The deep neural networks (DNN) that can replicate the behavior of the human expert who recognizes the characteristics of ECG waveform have been developed and studied to analyze ECG. However, although the existing DNNs can not provide the explanations for their decisions, those trials have attempted to determine whether patients have certain diseases or not and those decisions could not be accepted because of the absence of relating theoretical basis. In addition, these DNNs required a lot of training data to obtain sufficient accuracy in spite of the difficulty in the acquisition of relating clinical data. In this study, a small-sized continuous data processing DNN (C-DNN) was suggested to determine the simple characteristics of ECG wave that were not required additional explanations about its decisions and the C-DNN can be easily trained with small training data. Although it can analyze small input data that was selected in narrow region on whole ECG, it can continuously scan all ECG data and find important points such as start and end points of P, QRS and T waves within a short time. The star and end points of ECG waves determined by the C-DNNs were compared with the results performed by human experts to estimate the accuracies of the C-DNNs. The C-DNN has 150 inputs, 51 outputs, two hidden layers and one output layer. To find the start and end points, two C-DNNs were trained through deep learning technology and applied to a parameter acquisition algorithms. 12 lead ECG data measured in four patients and obtained through PhysioNet was processed to make training data by human experts. The accuracy of the C-DNNs were evaluated with extra data that were not used at deep learning by comparing the results between C-DNNs and human experts. The averages of the time differences between the C-DNNs and experts were 0.1 msec and 13.5 msec respectively and those standard deviations were 17.6 msec and 15.7 msec. The final step combining the results of C-DNN through the waveforms of 12 leads was successfully determined all 33 waves without error that the time differences of human experts decision were over 20 msec. The reliable decision of the ECG wave's start and end points benefits the acquisition of accurate ECG parameters such as the wave lengths, amplitudes and intervals of P, QRS and T waves.

• Journal of Periodontal and Implant Science
• /
• v.44 no.2
• /
• pp.79-84
• /
• 2014

Purpose: While single-species biofilms have been studied extensively, we know notably little regarding multispecies biofilms and their interactions. The purpose of this study was to develop and evaluate an in vitro multispecies dental biofilm model that aimed to mimic the environment of chronic periodontitis. Methods: Streptococcus gordonii KN1, Fusobacterium nucleatum ATCC23726, Aggregatibacter actinomycetemcomitans ATCC33384, and Porphyromonas gingivalis ATCC33277 were used for this experiment. The biofilms were grown on 12-well plates with a round glass slip (12 mm in diameter) with a supply of fresh medium. Four different single-species biofilms and multispecies biofilms with the four bacterial strains listed above were prepared. The biofilms were examined with a confocal laser scanning microscope (CLSM) and scanning electron microscopy (SEM). The minimum inhibitory concentrations (MIC) for four different planktonic single-species and multispecies bacteria were determined. The MICs of doxycycline and chlorhexidine for four different single-species biofilms and a multispecies biofilm were also determined. Results: The CLSM and SEM examination revealed that the growth pattern of the multispecies biofilm was similar to those of single-species biofilms. However, the multispecies biofilm became thicker than the single-species biofilms, and networks between bacteria were formed. The MICs of doxycycline and chlorhexidine were higher in the biofilm state than in the planktonic bacteria. The MIC of doxycycline for the multispecies biofilm was higher than were those for the single-species biofilms of P. gingivalis, F. nucleatum, or A. actinomycetemcomitans. The MIC of chlorhexidine for the multispecies biofilm was higher than were those for the single-species biofilms of P. gingivalis or F. nucleatum. Conclusions: To mimic the natural dental biofilm, a multispecies biofilm composed of four bacterial species was grown. The 24-hour multispecies biofilm may be useful as a laboratory dental biofilm model system.