• 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.

• Progress in Medical Physics
• /
• v.17 no.4
• /
• pp.212-223
• /
• 2006

Positional accuracy of the on-board imager (OBI) isocenter with gantry rotation was presented in this paper. Three different type of automatic evaluation methods of discrepancies between therapeutic and OBI isocenter using digital image processing techniques as well as a procedure stated in the customer acceptance procedure (CAP) were applied to check OBI isocenter migration trends. Two kinds of kV x-ray image set obtained at OBI source angle of $0^{\circ},\;90^{\circ},\;180^{\circ},\;270^{\circ}$ and every $10^{\circ}$ and raw projection data for cone-beam CT reconstruction were used for each evaluation method. Efficiencies of the methods were also estimated. If a user needs to obtain an isocenter variation map with full gantry rotation, a method taking OBI image for every $10^{\circ}$ and fitting with 5th order polynomial was appropriate. However for a mere quality assurance (QA) purpose of OBI isocenter accuracy, it was adequate to use only four OBI Images taken at the OBI source angle of $0^{\circ},\;90^{\circ},\;180^{\circ}\;and\;270^{\circ}$. Maximal discrepancy was 0.44 mm which was observed between the OBI source angle of $90^{\circ}\;and\;180^{\circ}$ OBI isocenter accuracy was maintained below 0.5 mm for a year. Proposed QA program may be helpful to Implement a reasonable routine QA of the OBI isocenter accuracy without great efforts.

• The KIPS Transactions:PartB
• /
• v.14B no.4
• /
• pp.311-320
• /
• 2007

This paper presents vision-based 3D facial expression animation technique and system which provide the robust 3D head pose estimation and real-time facial expression control. Many researches of 3D face animation have been done for the facial expression control itself rather than focusing on 3D head motion tracking. However, the head motion tracking is one of critical issues to be solved for developing realistic facial animation. In this research, we developed an integrated animation system that includes 3D head motion tracking and facial expression control at the same time. The proposed system consists of three major phases: face detection, 3D head motion tracking, and facial expression control. For face detection, with the non-parametric HT skin color model and template matching, we can detect the facial region efficiently from video frame. For 3D head motion tracking, we exploit the cylindrical head model that is projected to the initial head motion template. Given an initial reference template of the face image and the corresponding head motion, the cylindrical head model is created and the foil head motion is traced based on the optical flow method. For the facial expression cloning we utilize the feature-based method, The major facial feature points are detected by the geometry of information of the face with template matching and traced by optical flow. Since the locations of varying feature points are composed of head motion and facial expression information, the animation parameters which describe the variation of the facial features are acquired from geometrically transformed frontal head pose image. Finally, the facial expression cloning is done by two fitting process. The control points of the 3D model are varied applying the animation parameters to the face model, and the non-feature points around the control points are changed by use of Radial Basis Function(RBF). From the experiment, we can prove that the developed vision-based animation system can create realistic facial animation with robust head pose estimation and facial variation from input video image.

• The KIPS Transactions:PartB
• /
• v.19B no.2
• /
• pp.123-126
• /
• 2012

This paper proposed a method to separate a liver into left and right liver lobes for simple and exact volumetry of the river graft at abdominal MDCT(Multi-Detector Computed Tomography) image before the living donor liver transplantation. A medical team can evaluate an accurate river graft with minimized interaction between the team and a system using this algorithm for ensuring donor's and recipient's safe. On the image of segmented liver, 2 points(PMHV: a point in Middle Hepatic Vein and PPV: a point at the beginning of right branch of Portal Vein) are selected to separate a liver into left and right liver lobes. Middle hepatic vein is automatically segmented using PMHV, and the cutting line is decided on the basis of segmented Middle Hepatic Vein. A liver is separated on connecting the cutting line and PPV. The volume and ratio of the river graft are estimated. The volume estimated using 2 points are compared with a manual volume that diagnostic radiologist processed and estimated and the weight measured during surgery to support proof of exact volume. The mean ${\pm}$ standard deviation of the differences between the actual weights and the estimated volumes was $162.38cm^3{\pm}124.39$ in the case of manual segmentation and $107.69cm^3{\pm}97.24$ in the case of 2 points method. The correlation coefficient between the actual weight and the manually estimated volume is 0.79, and the correlation coefficient between the actual weight and the volume estimated using 2 points is 0.87. After selection the 2 points, the time involved in separation a liver into left and right river lobe and volumetry of them is measured for confirmation that the algorithm can be used on real time during surgery. The mean ${\pm}$ standard deviation of the process time is $57.28sec{\pm}32.81$ per 1 data set ($149.17pages{\pm}55.92$).

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.35 no.3
• /
• pp.366-372
• /
• 2006

This study is to investigate the physicochemical properties of differently pretreated chestnut starches during starch isolation and to examine their gelatinization properties by both heat and alkali treatments. One kind is starch A made by alkali method from peeled chestnut. The other is starch B made from chestnut with the outer layer. The results are as follows. Starch A has higher water binding capacity of 86.9% than starch B with 80.66%. Swelling powers of both starch A and B increased rapidly from $60^{\circ}C\;to\;80^{\circ}C$ in both, and since then it has changed a bit. Both began to show their solubility at $60^{\circ}C$ and increased continuously as the temperature went up. Starch A has higher swelling power and solubility than starch B. In iodine reaction, starch A has higher ${\lambda}max$ and absorbance at ${\lambda}max$ than starch B. X-ray diffraction patterns showed that starch A is type $C_b$ and that starch B is type B. Starch B has higher relative crystallinity of 37.0% than starch A with 36.2%. The results by differential scanning calorimetry revealed that starch A gelatinized from $66.95^{\circ}C$ to $77.5^{\circ}C$ and its enthalpy is 2.04 cal/g. And starch B gelatinized from $67.09^{\circ}C\;to\;77.5^{\circ}C$, and its enthalpy is 2.29 cal/g. Amylograms of chestnut starch at 6.5% concentration indicated that starch B needs higher onset temperature when beginning to gelatinize than starch A does. But starch A shows much higher peak viscosity, breakdown and setback than starch B does. Starch A shows higher viscosity, gel volume, and optical transmittance in gelatinization properties by alkali than starch B does.

• Journal of Chest Surgery
• /
• v.43 no.1
• /
• pp.11-19
• /
• 2010

Background: The commercially used vascular xenografts have some problems such as calcification, fibrosis and tissue degeneration that are associated with inflammatory and immunologic reactions. We compared two methods of xenograft preservation (fresh cryopreservation versus acellularized cryopreservation) of goat aorta. Material and Method: Aortic valved xenografts were harvested from adult pigs, and these were preserved using fresh cryopreservation (FC group, n=4) or acellularized crypreservation (AC group, n=4). These xenografts were implanted into adult goats. There were 2 short-term survivors (less than 100 days) and 2 long-term survivors in each group. These xenografts were explanted and they underwent microscopic examination. Result: The goats survived 31, 40, 107 and 411 days in the FC group and the other goats survived 5, 40, 363 and 636 days in the AC group. All the short-term survivors in each group expired because of rupture at the proximal anastomosis site. Marked neutrophil infiltration was observed in the FC group FC and lymphocytes were observed in the AC group. There were no differences in the occurrence of calcification, fibrosis and thrombosis among the groups. Conclusion: Some goats survived more than 100 days after the xenograft implantation irrespective of the methods of preservation. Because severe tissue degeneration developed in both groups, we think these methods are not appropriate for xenograft preservation of aorta. It was worth a preliminary trial for improving the preservation method or to modify the processing of xenografts.

• Journal of Food Hygiene and Safety
• /
• v.27 no.1
• /
• pp.103-107
• /
• 2012

We compared between an automated most-probable-number technique $TEMPO^{(R)}$TVC and traditional plating methods $Petrifilm^{TM}$ for estimating populations of total aerobic bacteria in various livestock products. 257 samples randomly selected in local retail stores and 87 samples inoculated with $E.$ $coli$ ATCC 25922, $Staphylococcus$ $aureus$ ATCC 12868 were tested in this study. The degree of agreement was estimated according to the CCFRA (Campden and Chorleywood Food Research Association Group) Guideline 29 and the agreement indicates the difference of two kinds methods is lower than 1 log base 10($log_{10}$). The samples of hams, jerky products, ground meat products, milks, ice creams, infant formulas, and egg heat formed products were showed above 95% in the agreement of methods. In contrast, proportion of agreement on meat extract products, cheeses and sausages were 93.1%, 92.1%, 89.1%, respectively. One press ham and five sausages containing spice and seasoning, two pork cutlets containing spice and bread crumbs, two meat extract product and two natural cheeses and one processing cheese with a high fat content, and one ice cream containing chocolate of all samples showed the discrepancy. Our result suggest that $TEMPO^{(R)}$TVC system is efficient to analyses total aerobic bacteria to compare manual method in time-consuming and laborious process except livestock products having limit of detection.

• Journal of Life Science
• /
• v.14 no.4
• /
• pp.650-656
• /
• 2004

Oryza grandiglumis (CCDD, 2n=48), one of the wild rice species, has been known to possess fungal-,bacterial-, and insect-resistance against sheath blight, rice blast, bacterial leaf blight and brown plant hopper (Nilaparvata lugens). To rapidly isolate differentially expressed genes responding to fungal and wounding stress, wounding and yeast extract were treated to O. grandiglumis for 24 hrs. Suppression subtractive hybridization (SSH) method was used to obtain differentially expressed genes from yeast extract and wounding treated plants. Seven hundreds and seventy six clones were obtained by subcloning PCR product, and colony array and screening were carried out using radio-isotope labeled cDNA probes prepared from the wounding and yeast extract treated plants. One hundred and fifteen colonies were confirmed as true positive ones. Average insert size of the clones were ranged from 400 bp to 700 bp and all the inserts were sequenced. To decide the identity of those clones, sequences were analyzed by sequence homology via GenBank database. The homology search result showed that 68 clones were matched to the genes with known function; 16 were related to primary metabolism, 5 to plant retrotransposons, 5 to defense related metallothionein-like genes. In addition to that, others were matched to various genes with known function in amino acid synthesis and processing, membrane transport, and signal transduction, so on. In northern blot analysis, induced expressions of ogwfi-161, ogwfi-646, ogwfi-663, and ogwfi-695 by wounding and yeast extract treatments were confirmed. The result indicates that SSH method is very efficient for rapid screening of differentially expressed genes.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.62 no.3
• /
• pp.184-192
• /
• 2017

Rice flours from five rice (Oryza sativa L.) varieties with different amylose content were prepared by both wet and dry milling processes. The moisture content of wet-milled rice flours (WMR) was approximately three-times higher than that of dry-milled rice flours (DMR). Water absorption index (WAI), water solubility index (WSI), and swelling power (SP) increased in proportion to temperature. The WAI, WSI and SP values of DMR were higher than those of WMR. Baeokchal (BOC), which is a waxy rice cultivar, had a significantly high WSI value. Pasting properties of DMR, except for the BOC cultivar, resulted in an increase in peak, trough, final, and setback viscosities. The levels of resistant starch in four cultivars, except for Dodamssal (DDS), were under 1%, irrespective of the milling process, whereas the resistant starch contents of DMR and WMR in DDS were 9.18% and 6.27%, respectively. In vitro digestibility of WMR was higher than that of DMR, and the estimated glycemic index of the rice flour varieties ranged from 57.6 to 81.3. Damaged starch content of WMR was less than that of DMR; in addition, a negative correlation was observed between the amylose and damaged starch contents of WMR. These results suggest that the properties of rice flour vary depending on the milling method and flour variety, and could be a reference for selecting the appropriate processing method.

• Korean Journal of Food Science and Technology
• /
• v.27 no.3
• /
• pp.281-285
• /
• 1995

The persimmon leaf tea was produced from persimmon leaves by three different methods (conventional, steamed, fermented) and the changes of total vitamin C and superoxide dismutase(SOD)-like activity were investigated. The amount of total vitamin C was the highest in fermented persimmon leaf tea which was 47% of total vitamin C of raw persimmon leaves. The SOD-like activity of conventional and fermented persimmon leaf teas were a little higher than that of steamed. The total vitamin C of steamed persimmon leaf tea was decreased slowly as the extraction time was increased. That of fermented persimmon leaf tea was increased generally at $60^{\circ}C\;and\;70^{\circ}C$, and increased until 5min and then decreased at $80^{\circ}C\;and\;90^{\circ}C$, The optimum extraction condition of total vitamin C in fermented persimmon leaf tea was 3min and 5min at $80^{\circ}C\;and\;90^{\circ}C$, 5min and 10min at $60^{\circ}C$. The total vitamin C of conventional persimmon leaf tea was so little that could not be measured by DNP method. The SOD-like activity of conventional and steamed persimmon leaf teas were increased and that of fermented showed the trend of increasing-decreasing-increasing. Fermented persimmon leaf tea had higher SOD-like activity than conventional and steamed at all tested conditions, and the optimum extraction condition of SOD-like activity was 3min and 10min at $80^{\circ}C\;and\;90^{\circ}C$.