• 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 the Korea Academia-Industrial cooperation Society
• /
• v.18 no.12
• /
• pp.126-134
• /
• 2017

There have been numerous studies on extracting the R-peak from electrocardiogram (ECG) signals. However, most of the detection methods are complicated to implement in a real-time portable electrocardiograph device and have the disadvantage of requiring a large amount of calculations. R-peak detection requires pre-processing and post-processing related to baseline drift and the removal of noise from the commercial power supply for ECG data. An adaptive filter technique is widely used for R-peak detection, but the R-peak value cannot be detected when the input is lower than a threshold value. Moreover, there is a problem in detecting the P-peak and T-peak values due to the derivation of an erroneous threshold value as a result of noise. We propose a robust R-peak detection algorithm with low complexity and simple computation to solve these problems. The proposed scheme removes the baseline drift in ECG signals using an adaptive filter to solve the problems involved in threshold extraction. We also propose a technique to extract the appropriate threshold value automatically using the minimum and maximum values of the filtered ECG signal. To detect the R-peak from the ECG signal, we propose a threshold neighborhood search technique. Through experiments, we confirmed the improvement of the R-peak detection accuracy of the proposed method and achieved a detection speed that is suitable for a mobile system by reducing the amount of calculation. The experimental results show that the heart rate detection accuracy and sensitivity were very high (about 100%).

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.3 no.4
• /
• pp.183-192
• /
• 1998

The long-term observation of temperature in the coastal waters adjacent to the Wolsung Nuclear Power Plant has been carried out from November 10, 1996 to August 22, 1997, for approximately 280 days using a real-time temperature measurement buoy system. The sea-surface temperature was measured at every 10 minute using 10 buoys. The vertical structure of temperature was investigated near the outlet of the plant with two thermistor chains equipped with 10 sensors at 1 m interval The monthly averaged temperature was the lowest with spatial average of $12.8^{\circ}C$ in February and was the highest in August with spatial average of $19.6^{\circ}C$. The extremely low temperature was frequently observed between June and August, which seems to be the consequence of the intrusion of cold water near the southeastern coast of Korea. Distributions of the daily and hourly averaged temperature show that the highest temperature always occurred near the outlet of the plant and the warm-water patch moved along the north-south direction with the semidiurnal period. The semidiurnal fluctuation of temperature was also observed near the surface of the vertical profiles. The spectral analysis of temperature between February and April 1997 shows that the semidiurnal components prevailed near the outlet. It is likely that the semidiurnal components were due to the prevailing semidiurnal tide in this region. In August 1997, the diurnal components were dominant at the surface water of all stations except Station 12, which suggests that the warm water from the outlet of the plant has less effects in summer on the surrounding waters than the strong solar radiation.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.17 no.1
• /
• pp.25-34
• /
• 2015

In this research, we applied a procedure of quality control (QC) to the agro-meteorological data measured at the Suwon weather station of Korea Meteorological Administration (KMA). The QC was conducted through six steps based on the KMA Real-time Quality control system for Meteorological Observation Data (RQMOD) and four steps based on the International Soil Moisture Network (ISMN) QC modules. In addition, we set up our own empirical method to remove erroneous data which could not be filtered by the RQMOD and ISMN methods. After all these QC procedures, a well-refined agro-meteorological dataset was complied at both air and soil temperatures. Our research suggests that soil moisture requires more detailed and reliable grounds to remove doubtful data, especially in winter with its abnormal variations. The raw data and the data after QC are now available at the NCAM website (http://ncam.kr/page/req/agri_weather.php).

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.24 no.4
• /
• pp.69-79
• /
• 1982

The objective of this study is to contribute to drainage planning in the most realistic and economical way by establishing the relationship between rice yield reduction and overhead flooding by muddy water of each growth stage of paddy, which is the most important factor in determining optimum drainage facilities. This study was based on the data mainly from the experimental reports of the Office of Rural Development of Korea, Reduction Rate Estimation for Summer Crops, published by Ministry of Agriculture and Forestry of Japan and other related research documenta- tion. The results of this study are summarized as follows 1. Damages by overhead flooding are highest in heading stage and have the tendency of decrease in the order of booting stage, panicle formation stage, tillering stage, and stage just after transplanting. Damages by overhead flooding of each growing stage are as follows: a) It is considered that overhead flooding just after transplanting gives a little influence on plant growth and yield because the paddy has sufficient growth period from floo ding to harvest time. b) Jt is analyzed that according to the equation y=11 12x 0.908 which is derived from this study, damages by overhead flooding during tillering stage for 1, 2, 3 successive days are 11.1 %, 20.9%, and 30.2% respectively. c) Damages by overhead flooding after panicle formation stage are very serious because recovering period is very short after damage and ineffective tillering is much. Acc- ording to the equation y=9. 58x+10. Ol derived from this study, damages by overhead flooding fal 1,2,3,5 successive days are 19.6%, 29.2%, 38.8%, 57.9% respectively. d) Booting stage is the very important period in which young panicle has grown up almost completely and the number of glumous flower is fixed since reduction division takes place in the microspore mother cell and enbryo mother cell. According to the equation y=39. 66x 0.558 derived from this study, damages by overhead floodingfor 0.5, 1, 3, 5 successive days are 26.9%, 39.7%, 72. 2% and 97.4%, respectively. Therefore, damages by overhead flooding is very serious during the hooting stage. e) When ear of paddy emerges, flowering begins on that day or the next day; when paddy flowers, fertilization will be completed 2-3 hours after flowering. Therefore overhead flooding during heading stage impedes flowering and increases sterilizing percentage. From this reason damages of heading stage are larger than that of booting stage. According to the equation y-41 94x 0.589 derived from this study, damages by overhead flooding for 0.5, 1, 3, 5, successive days are 27.9%, 63.1 %, 80.1%, and 100% 2. Considering that temperature of booting stage is higher than that of beading stage and plant height of booting stage is ten centimeters shorter than that of heading stage, booting stage should be taken as a critical period for drainage planning because possi- bility of damage occurrence in booting stage is larger than that of heading stage. There-fore, it is considered that booting stage should be taken as critical period of paddy growth for drainage planning. 3. Overhead flooding depth is different depending on the stage of growth. In case, booting stage is adopted as design stage of growth for drainage planning, it is conside red that the allowable flooding depth for new varieties and general varieties are 70cm and 80cm respectively. 4. Reduction Rate Estimation by Wind and Flood for Rice Planting of the present design criteria for drainage planning shows damage by overhead flooding for 1 to 2, 3 to 4, 5 to 7 consecutive days; damages by overhead flooding varies considerably over several hours and experimental condition of soil, variety of paddy, and climate differs with real situation. From these reasons, damage by flooding could not be estimated properly in the past. This study has derived the equation which shows damages by flooding of each growth stage on an hourly basis. Therefore, it has become possible to compute the exact damages in case duration of overhead flooding is known.

• Pediatric Infection and Vaccine
• /
• v.18 no.2
• /
• pp.182-192
• /
• 2011

Purpose : Atopic findings may be associated with severity of pneumonia in 2009 pandemic influenza A (H1N1) infection, which could suggest a possible association between atopic findings and the severity of viral infections. Thus, we studied association between atopic findings and severity of disease in children with H1N1 influenza infection. Methods : A retrospective study was performed in 74 children admitted in a single tertiary institute and confirmed as H1N1 patients by reverse transcriptase (RT) - polymerase chain reaction (PCR). They were divided into 2 groups according to the severity of pneumonia. We evaluated whether the atopic finding is risk factor between the two groups. Results : Children with severe pneumonia had higher percentages of serum eosinophilia (88% vs 40%, P <0.001), asthma (65% vs 35%, P =0.011), allergic rhinitis (71% vs 40%, P =0.009), and IgE level (P =0.007). We found positive correlations between aeroallergen sensitizations and severity of pneumonia (82% vs 53%, P =0.007). Conclusion : Among patients with H1N1 pneumonia, asthma and atopic findings are risk factors for severity of pneumonia.

• Journal of Korean Philosophical Society
• /
• v.142
• /
• pp.1-25
• /
• 2017

Kant's doctrine of virtue shows how an actor should behave morally in an individual situation with moral law defines the limits of human action. There is latitude for action in the course of formulating the maxims of action by an actor. And moral judgement, as Aristotle's Pronesis, is very important in the latitude for action. In the doctrine of virtue, Kant suggests two kinds of duty of virtue: one's own perfeciton as an obligatory end, and the happiness to others as an obligatory end-and raises the question of casuistics for each. However, this was the practice and training for the human moral life by application of the moral law. In particular, Kant saw that ethics does not give laws for action, but only give laws for the maxims of action, and further intended to realize the practice in a proper way of seeking truth through casuistical questions. Thus, Kant points out that the casuistic is related only to ethics in a fragmentary way and is added to ethics only as a comment on the system. According to Kant, virtue and judgment are inevitable to apply categorical imperative in the empirical and realistic world. In other words, virtue and judgment are necessary to enable people who are likely to act in accordance to inclination to live a moral life in accordance with the command of reason. Thus Kant saw that in order to take wide duty into narrow ones, human beings must not only have to cultivate virtues as a strong power of will, but also to exercise judgment. In addition, the distinction between duty of law(narrow obligation) and duty of virtue(wide obligation) is dependent on whether there is a latitude for action in the application of both duties. So the role of virtue and training of judgement is very important in the latitude for action that occurs in the process of formalizing actor's maxims. In detail, as the duty is wider, so man's obligation to action is more imperfect, but the closer to narrow duty(Law) he brings the maxim of observing this duty(in his attitude of will), so much the more perfect is his virtuous action. Thus, it was an effort to show how Kant's best moral principles, that is categorical imperative could be applied to the real world at the time of criticism. Of course, even if it is difficult to assess Kant's efforts as successful, criticizing Kant's ethics as 'formal', 'abstract', or 'monologous' is not persuasive because of critics did not understand his ethics as a whole.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.43-62
• /
• 2019

At one time, the anomaly detection sector dominated the method of determining whether there was an abnormality based on the statistics derived from specific data. This methodology was possible because the dimension of the data was simple in the past, so the classical statistical method could work effectively. However, as the characteristics of data have changed complexly in the era of big data, it has become more difficult to accurately analyze and predict the data that occurs throughout the industry in the conventional way. Therefore, SVM and Decision Tree based supervised learning algorithms were used. However, there is peculiarity that supervised learning based model can only accurately predict the test data, when the number of classes is equal to the number of normal classes and most of the data generated in the industry has unbalanced data class. Therefore, the predicted results are not always valid when supervised learning model is applied. In order to overcome these drawbacks, many studies now use the unsupervised learning-based model that is not influenced by class distribution, such as autoencoder or generative adversarial networks. In this paper, we propose a method to detect anomalies using generative adversarial networks. AnoGAN, introduced in the study of Thomas et al (2017), is a classification model that performs abnormal detection of medical images. It was composed of a Convolution Neural Net and was used in the field of detection. On the other hand, sequencing data abnormality detection using generative adversarial network is a lack of research papers compared to image data. Of course, in Li et al (2018), a study by Li et al (LSTM), a type of recurrent neural network, has proposed a model to classify the abnormities of numerical sequence data, but it has not been used for categorical sequence data, as well as feature matching method applied by salans et al.(2016). So it suggests that there are a number of studies to be tried on in the ideal classification of sequence data through a generative adversarial Network. In order to learn the sequence data, the structure of the generative adversarial networks is composed of LSTM, and the 2 stacked-LSTM of the generator is composed of 32-dim hidden unit layers and 64-dim hidden unit layers. The LSTM of the discriminator consists of 64-dim hidden unit layer were used. In the process of deriving abnormal scores from existing paper of Anomaly Detection for Sequence data, entropy values of probability of actual data are used in the process of deriving abnormal scores. but in this paper, as mentioned earlier, abnormal scores have been derived by using feature matching techniques. In addition, the process of optimizing latent variables was designed with LSTM to improve model performance. The modified form of generative adversarial model was more accurate in all experiments than the autoencoder in terms of precision and was approximately 7% higher in accuracy. In terms of Robustness, Generative adversarial networks also performed better than autoencoder. Because generative adversarial networks can learn data distribution from real categorical sequence data, Unaffected by a single normal data. But autoencoder is not. Result of Robustness test showed that he accuracy of the autocoder was 92%, the accuracy of the hostile neural network was 96%, and in terms of sensitivity, the autocoder was 40% and the hostile neural network was 51%. In this paper, experiments have also been conducted to show how much performance changes due to differences in the optimization structure of potential variables. As a result, the level of 1% was improved in terms of sensitivity. These results suggest that it presented a new perspective on optimizing latent variable that were relatively insignificant.

• Korean journal of applied entomology
• /
• v.58 no.2
• /
• pp.111-120
• /
• 2019

Riptortus pedestris (Fabricius) (Hemiptera: Alydidae) is an economically important insect pest of soybean and fruit trees. We investigated the temperature effects on the adult fecundity and longevity, and determined the parameters of oviposition models and life table at different constant temperatures 15.8, 19.7, 24.0, 27.8, 32.6, 34.0, and $35.5^{\circ}C$. R. pedestris females reproduced successfully from 19.7 to $35.5^{\circ}C$ except $15.8^{\circ}C$. The longevity of R. pedestris was longest at $15.8^{\circ}C$ and it decreased with increasing temperature (76.6 days at $19.7^{\circ}C$ and 20.6 days at $35.5^{\circ}C$). The number of total eggs and viable eggs was highest at $24.0^{\circ}C$ (193.5 and 151.2). Egg hatchability was highest at $27.8^{\circ}C$ (84.0%). We compared the results of oviposition models and life table parameters using both total eggs and viable eggs. The parameter value (c: the maximum reproductive capacity) (190 eggs) of temperature dependent total fecundity model using total eggs was higher than that of the model using viable eggs. When we analyzed the life table parameter the values of net reproductive rate and mean generation time using viable eggs were lower than those using total eggs. The oviposition models and life table analysis using viable eggs will be helpful to understand the real population transition of R. pedestris in agricultural system.

• Journal of the Korean Society of Radiology
• /
• v.15 no.1
• /
• pp.45-54
• /
• 2021

This study was written to create educational content in non-destructive fields based on Mixed Reality. Currently, in the field of radiation, there is almost no content for educational Mixed Reality-based educational content. And in the field of non-destructive inspection, the working environment is poor, the number of employees is often 10 or less for each manufacturer, and the educational infrastructure is not built. There is no practical training, only practical training and safety education to convey information. To solve this, it was decided to develop non-destructive worker education content based on Mixed Reality. This content was developed based on Microsoft's HoloLens 2 HMD device. It is manufactured based on the resolution of 1280 ⁎ 720, and the resolution is different for each device, and the Side is created by aligning the Left, Right, Bottom, and TOP positions of Anchor, and the large image affects the size of Atlas. The large volume like the wallpaper and the upper part was made by replacing it with UITexture. For UI Widget Wizard, I made Label, Buttom, ScrollView, and Sprite. In this study, it is possible to provide workers with realistic educational content, enable self-directed education, and educate with 3D stereoscopic images based on reality to provide interesting and immersive education. Through the images provided in Mixed Reality, the learner can directly operate things through the interaction between the real world and the Virtual Reality, and the learner's learning efficiency can be improved. In addition, mixed reality education can play a major role in non-face-to-face learning content in the corona era, where time and place are not disturbed.