• Journal of Intelligence and Information Systems
• /
• v.18 no.2
• /
• pp.29-45
• /
• 2012

Bond rating is regarded as an important event for measuring financial risk of companies and for determining the investment returns of investors. As a result, it has been a popular research topic for researchers to predict companies' credit ratings by applying statistical and machine learning techniques. The statistical techniques, including multiple regression, multiple discriminant analysis (MDA), logistic models (LOGIT), and probit analysis, have been traditionally used in bond rating. However, one major drawback is that it should be based on strict assumptions. Such strict assumptions include linearity, normality, independence among predictor variables and pre-existing functional forms relating the criterion variablesand the predictor variables. Those strict assumptions of traditional statistics have limited their application to the real world. Machine learning techniques also used in bond rating prediction models include decision trees (DT), neural networks (NN), and Support Vector Machine (SVM). Especially, SVM is recognized as a new and promising classification and regression analysis method. SVM learns a separating hyperplane that can maximize the margin between two categories. SVM is simple enough to be analyzed mathematical, and leads to high performance in practical applications. SVM implements the structuralrisk minimization principle and searches to minimize an upper bound of the generalization error. In addition, the solution of SVM may be a global optimum and thus, overfitting is unlikely to occur with SVM. In addition, SVM does not require too many data sample for training since it builds prediction models by only using some representative sample near the boundaries called support vectors. A number of experimental researches have indicated that SVM has been successfully applied in a variety of pattern recognition fields. However, there are three major drawbacks that can be potential causes for degrading SVM's performance. First, SVM is originally proposed for solving binary-class classification problems. Methods for combining SVMs for multi-class classification such as One-Against-One, One-Against-All have been proposed, but they do not improve the performance in multi-class classification problem as much as SVM for binary-class classification. Second, approximation algorithms (e.g. decomposition methods, sequential minimal optimization algorithm) could be used for effective multi-class computation to reduce computation time, but it could deteriorate classification performance. Third, the difficulty in multi-class prediction problems is in data imbalance problem that can occur when the number of instances in one class greatly outnumbers the number of instances in the other class. Such data sets often cause a default classifier to be built due to skewed boundary and thus the reduction in the classification accuracy of such a classifier. SVM ensemble learning is one of machine learning methods to cope with the above drawbacks. Ensemble learning is a method for improving the performance of classification and prediction algorithms. AdaBoost is one of the widely used ensemble learning techniques. It constructs a composite classifier by sequentially training classifiers while increasing weight on the misclassified observations through iterations. The observations that are incorrectly predicted by previous classifiers are chosen more often than examples that are correctly predicted. Thus Boosting attempts to produce new classifiers that are better able to predict examples for which the current ensemble's performance is poor. In this way, it can reinforce the training of the misclassified observations of the minority class. This paper proposes a multiclass Geometric Mean-based Boosting (MGM-Boost) to resolve multiclass prediction problem. Since MGM-Boost introduces the notion of geometric mean into AdaBoost, it can perform learning process considering the geometric mean-based accuracy and errors of multiclass. This study applies MGM-Boost to the real-world bond rating case for Korean companies to examine the feasibility of MGM-Boost. 10-fold cross validations for threetimes with different random seeds are performed in order to ensure that the comparison among three different classifiers does not happen by chance. For each of 10-fold cross validation, the entire data set is first partitioned into tenequal-sized sets, and then each set is in turn used as the test set while the classifier trains on the other nine sets. That is, cross-validated folds have been tested independently of each algorithm. Through these steps, we have obtained the results for classifiers on each of the 30 experiments. In the comparison of arithmetic mean-based prediction accuracy between individual classifiers, MGM-Boost (52.95%) shows higher prediction accuracy than both AdaBoost (51.69%) and SVM (49.47%). MGM-Boost (28.12%) also shows the higher prediction accuracy than AdaBoost (24.65%) and SVM (15.42%)in terms of geometric mean-based prediction accuracy. T-test is used to examine whether the performance of each classifiers for 30 folds is significantly different. The results indicate that performance of MGM-Boost is significantly different from AdaBoost and SVM classifiers at 1% level. These results mean that MGM-Boost can provide robust and stable solutions to multi-classproblems such as bond rating.

• The Journal of Korean Society for Radiation Therapy
• /
• v.12 no.1
• /
• pp.91-104
• /
• 2000

Recently linear accelerator in radiation therapy in asymmetric field has been easily used since the improvement and capability of asymmetrical field adjustment attached to the machine. It has been thought there have been some significant errors in dose calculation when asymmetrical radiation fields have been utilized in practice of radiation treatments if the fundamental data for dose calculation have been measured in symmetrical standard fields. This study investigated how much the measured data of dose distributions and their isodose curves are different between in asymmetrical and symmetrical standard fields, and how much there difference affect the error in dose calculation in conventional method measured in symmetrical standard field. The distributions of radiation dose were measured by photon diode detector in the water phantom (RFA-300P, Scanditronix, Sweden) as tissue equivalent material on utilization of 6 MV linear accelerator with source surface distance (SSD) 1000 mm. The photon diode detector has the velocity of 1 mm per second from water surface to 250 mm depth in the field size of $40mm{\times}40mm\;to\;250mm{\times}250mm\;symmetric\;field\;and\;40mm{\times}20mm\;to\;250mm{\times}125mm$ asymmetrical fields. The measurements of percent depth dose (PDD) and subsequent plotting of their isodose curves were performed from water surface to 250mm dmm from Y-center axis in $100mm{\times}50mm$ field in order to absence the variability of depth dose according to increasing field sizes and their affects to plotted isodose curves. The difference of PDD between symmetric and asymmetric field was maximum $4.1\%\;decrease\;in\;40mm{\times}20mm\;field,\;maximum\;6.6\%\;decrease\;in\;100mm{\times}50mm\;and\;maximum\;10.2\%\;decrease\;200mm{\times}100mm$, the larger decrease difference of PDD as the greater field size and as greater the depth, The difference of PDD between asymmetrical field and equivalent square field showed maximum $2.4\%\;decrease\;in\;60mm{\times}30mm\;field,\;maximum\;4.8\%\;decrease\;in\;150mm{\times}75mm\;and\;maximum\;6.1\%\;decrease\;in\;250mm{\times}125mm$, and the larger decreased differenced PDD as the greater field size and as greater the depth, these differences of PDD were out of $5\%$ of dose calculation as defined by international Commission on radiation unit and Measurements(ICRU). In the dose distribution of asymmetrical field (half beam) the plotted isodose curves were observed to have deviations by decreased PDD as greater as the blocking of the beam moved closer to the central axis, and as the asymmetrical field increased by moving the block 10 mm keeping away from the central axis, the PDD increased and plotted isodose curves were gradually more flattened, due to reduced amount of the primary beam and the fraction of low energy soft radiations by passing thougepth in asymmetrical field by moving independent jaw each 10 h beam flattening filter. As asymmetrical radiation field as half beam radiation technique is used, the radiation dosimetry calculated in utilizing the fundamental data which measured in standard symmetrical field should be converted on bases of nearly measured data in asymmetrical field, measured beam data flies of various asymmetrical field in various energy and be necessary in each institution.

• Fashion & Textile Research Journal
• /
• v.6 no.1
• /
• pp.41-50
• /
• 2004

This research proposes the on-stage costumes for the play Hamlet of Shakespeare performed by Yunheedan Guhri Pae - the Street Theater Troupe. Stage costumes have an important role in displaying the characteristics of each characters to the audience and has big visual effects. However, in order to design the costumes in the object viewpoints of the audience, the survey on the images of the characters who had actually watched the performance was taken place and proposed the costume design according to the results of the survey. Hamlet a: This result was applied to propose a sweater in black color, black leather pants and vest. Hamlet b: This result was applied to propose hooded coat in purple in middle level of brightness and color spectrum and yellow coat. For free image, loose pants in blue and vest in the same color tone were proposed. Gertrude a: This result was applied to use purple (violet) with reddish tone to propose the formation of a dress applying tailored suit. Gertrude b: This result was applied to propose purple gown and the one-piece dress with black laces. Ophelia a: This result was applied to propose feminine white dress and cape in purple color tone. Ophelia b: This result was applied to propose dyed and weaved clothes. Through the surveys as above, the images of each character was driven in adjectives, and using the results driven from the brightness, coloration, and color, color images were proposed. Only one costume cannot make up for the stage costumes and because it exists as an element of stage production, it is true that costumes are limited in some areas. However, that limit can become the motive of the costume. There is a limit, which the designer cannot produce the costumes as he or she had designed but I believe it is the center of the on-stage customers to display the characteristics of the characters according to the given concept. The limit of this research is the fact that because the costumes were designed so they fit the conditions already given, thus it was difficult to regard the process of designing and producing the costume as a project done according to the interaction. And in the future, if it is possible, I wish for the joint research with the people responsible for stage art to take place as a practical stage art. It was possible to produce practical costume since they were produced for actual performance and the production of costumes considering the dance steps, line of flow, and acting, was able to reduce the trial and error on stage. Through this research, I felt that the understanding and smooth interaction on diverse other areas not limited to the costume design should be taken place and believe that this was a research that proposes new research method since there had been only a few previous research regarding the on-stage costumes for actual performances. Therefore, this research had depended on the surveys given to the audiences to endow objectivity, however, I wish this research can contribute to defining effective process and methods for the on-stage costumes with more active researches with diverse methods and in diverse areas. I am sorry that the costume production for all the characters and all the scenes in Hamlet couldn't be done due to many limitations. As the following research assignment, I am planning on designing the costumes for all the scenes.

• The Journal of Korean Society for Radiation Therapy
• /
• v.26 no.2
• /
• pp.207-216
• /
• 2014

Purpose : We present a method to reduce this gap and complete the treatment plan, to be made by the re-optimization is performed in the same conditions as the initial treatment plan different from Monaco treatment planning system. Materials and Methods : The optimization is carried in two steps when performing the inverse calculation for volumetric modulated radiation therapy or intensity modulated radiation therapy in Monaco treatment planning system. This study was the first plan with a complete optimization in two steps by performing all of the treatment plan, without changing the optimized condition from Step 1 to Step 2, a typical sequential optimization performed. At this time, the experiment was carried out with a pencil beam and Monte Carlo algorithm is applied In step 2. We compared initial plan and re-optimized plan with the same optimized conditions. And then evaluated the planning dose by measurement. When performing a re-optimization for the initial treatment plan, the second plan applied the step optimization. Results : When the common optimization again carried out in the same conditions in the initial treatment plan was completed, the result is not the same. From a comparison of the treatment planning system, similar to the dose-volume the histogram showed a similar trend, but exhibit different values that do not satisfy the conditions best optimized dose, dose homogeneity and dose limits. Also showed more than 20% different in comparison dosimetry. If different dose algorithms, this measure is not the same out. Conclusion : The process of performing a number of trial and error, and you get to the ultimate goal of treatment planning optimization process. If carried out to optimize the completion of the initial trust only the treatment plan, we could be made of another treatment plan. The similar treatment plan could not satisfy to optimization results. When you perform re-optimization process, you will need to apply the step optimized conditions, making sure the dose distribution through the optimization process.

• Korean Journal of Agricultural Science
• /
• v.8 no.2
• /
• pp.195-202
• /
• 1981

The return of water to the atmosphere from water, soil and vegetation surface is one of the most important aspects of hydrological cycle, and the seasonal trend of variation of river basin evaporation is also meaningful in the longterm runoff analysis for the irrigation and water resources planning. This paper has been prepared to show some imformation to estimate the monthly river basin evaporation from pan evaporation, potential evaporation, regional evaporation and temperature through the comparison with river basin evaporation derived from water budget method. The analysis has been carried out with the observation data of Yongdam station in the Geum river basin for five year. The results are summarized as follows and these would be applied to the estimation of river basin evaporation and longterm runoff in ungaged station. 1. The ratio of pan evaporation to river basin evaporation ($E_w/E_{pan}$) shows the most- significant relation at the viewpoint of seasonal trend of variation. River basin evaporation could be estimated from the pan evaporation through either Fig. 9 or Table-7. 2. Local coefficients of cloudness effect and wind function has been determined to apply the Penman's mass and energy transfer equation to the estimation of river basin evaporation. $R_c=R_a(0.13+0.52n/D)$ $E=0.35(e_s-e)(1.8+1.0U)$ 3. It seems that Regional evaporation concept $E_R=(1-a)R_C-E_p$ has kept functional errors due to the inapplicable assumptions. But it is desirable that this kind of function which contains the results of complex physical, chemical and biological processes of river basin evaporation should be developed. 4. Monthly river basin evaporation could be approximately estimated from the monthly average temperature through either the equation of $E_w=1.44{\times}1.08^T$ or Fig. 12 in the stations with poor climatological observation data.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.16 no.2
• /
• pp.199-210
• /
• 2005

Objective : Attention deficit hyperactivity disorder(ADHD) affects $5-10\%$ of children in Korea, with more boys and girls being diagnosed. Despite seriousness of ADHD, little is known about its causes. From the current genetic epidemiologic studies, ADHD is known as a heritable disorder. Till now, however, there have been very few genetic studies about ADHD in Korea. The aim of the this study is to examine the association between dopamine transporter gone type 1 and ADHD using case-control design in Korean ADHD probands and normal controls. Materials and Method : Child Psychiatric Genetic research team in Seoul National University Hospital, Clinical Research Institute recruited the ADHD probands using clinical interview/observation, diverse rating scales, and neuropsychological tests. For eliminating phenocopy or ADHD, diagnosis of ADHD was based upon clinical data, psychometric data, and parent/teacher reports. Total 85 ADHD-probands were recruited as final study subjects and independent 100 normal adults participated in this study as control group. For all the ADHD probands, and controls, the 3'-UTR-VNTR polymorphism of DAT1 was analyzed. Based on the DAT1 allele and genotype informations, Chi-square test based on case-control design was performed. Results : As for genetic study, total of 85 probands and 100 controls were included for the genetic analysis. Four different alleles, 350bp (7repeat), 440bp (9repeat), 480bp (10repeat) and 520bp (11repeat) were found in DAT1 gene of study subjects. In case-control analysis, ADHD probands and parents have significantly more 9 repeat allele and 9/10 genotype. Also, The probands with 9repeat allele have more commission errors in ADS. Conclusion : The positive association between ADHD and DAT1 gene was replicated in this report like other previous results for caucasian children and Korean children with ADHD. There are ongoing studies on other candidate genes such as DRD4 and DRD5 and it would be required to explore the association of these candidate genes in Korean children with ADHD. These ongoing genetic research will contribute to the understanding of heterogenous genetic and environmental etiologies of ADHD phenotype, which will lead to the development of more comprehensive treatment and preventive interventions for ADHD.

• Journal of Intelligence and Information Systems
• /
• v.22 no.3
• /
• pp.113-127
• /
• 2016

Among the techniques for recommendation, collaborative filtering (CF) is commonly recognized to be the most effective for implementing recommender systems. Until now, CF has been popularly studied and adopted in both academic and real-world applications. The basic idea of CF is to create recommendation results by finding correlations between users of a recommendation system. CF system compares users based on how similar they are, and recommend products to users by using other like-minded people's results of evaluation for each product. Thus, it is very important to compute evaluation similarities among users in CF because the recommendation quality depends on it. Typical CF uses user's explicit numeric ratings of items (i.e. quantitative information) when computing the similarities among users in CF. In other words, user's numeric ratings have been a sole source of user preference information in traditional CF. However, user ratings are unable to fully reflect user's actual preferences from time to time. According to several studies, users may more actively accommodate recommendation of reliable others when purchasing goods. Thus, trust relationship can be regarded as the informative source for identifying user's preference with accuracy. Under this background, we propose a new hybrid recommender system that fuses CF and social network analysis (SNA). The proposed system adopts the recommendation algorithm that additionally reflect the result analyzed by SNA. In detail, our proposed system is based on conventional memory-based CF, but it is designed to use both user's numeric ratings and trust relationship information between users when calculating user similarities. For this, our system creates and uses not only user-item rating matrix, but also user-to-user trust network. As the methods for calculating user similarity between users, we proposed two alternatives - one is algorithm calculating the degree of similarity between users by utilizing in-degree and out-degree centrality, which are the indices representing the central location in the social network. We named these approaches as 'Trust CF - All' and 'Trust CF - Conditional'. The other alternative is the algorithm reflecting a neighbor's score higher when a target user trusts the neighbor directly or indirectly. The direct or indirect trust relationship can be identified by searching trust network of users. In this study, we call this approach 'Trust CF - Search'. To validate the applicability of the proposed system, we used experimental data provided by LibRec that crawled from the entire FilmTrust website. It consists of ratings of movies and trust relationship network indicating who to trust between users. The experimental system was implemented using Microsoft Visual Basic for Applications (VBA) and UCINET 6. To examine the effectiveness of the proposed system, we compared the performance of our proposed method with one of conventional CF system. The performances of recommender system were evaluated by using average MAE (mean absolute error). The analysis results confirmed that in case of applying without conditions the in-degree centrality index of trusted network of users(i.e. Trust CF - All), the accuracy (MAE = 0.565134) was lower than conventional CF (MAE = 0.564966). And, in case of applying the in-degree centrality index only to the users with the out-degree centrality above a certain threshold value(i.e. Trust CF - Conditional), the proposed system improved the accuracy a little (MAE = 0.564909) compared to traditional CF. However, the algorithm searching based on the trusted network of users (i.e. Trust CF - Search) was found to show the best performance (MAE = 0.564846). And the result from paired samples t-test presented that Trust CF - Search outperformed conventional CF with 10% statistical significance level. Our study sheds a light on the application of user's trust relationship network information for facilitating electronic commerce by recommending proper items to users.

• Journal of the Korean earth science society
• /
• v.38 no.4
• /
• pp.283-292
• /
• 2017

In this study, the empirical models were established to estimate the concentrations of surface-level $PM_{2.5}$ over Seoul, Korea from 1 January 2012 to 31 December 2013. We used six different multiple linear regression models with aerosol optical thickness (AOT), ${\AA}ngstr{\ddot{o}}m$ exponents (AE) data from Moderate Resolution Imaging Spectroradiometer (MODIS) aboard Terra and Aqua satellites, meteorological data, and planetary boundary layer depth (PBLD) data. The results showed that $M_6$ was the best empirical model and AOT, AE, relative humidity (RH), wind speed, wind direction, PBLD, and air temperature data were used as input data. Statistical analysis showed that the result between the observed $PM_{2.5}$ and the estimated $PM_{2.5}$ concentrations using $M_6$ model were correlations (R=0.62) and root square mean error ($RMSE=10.70{\mu}gm^{-3}$). In addition, our study show that the relation strongly depends on the seasons due to seasonal observation characteristics of AOT, with a relatively better correlation in spring (R=0.66) and autumntime (R=0.75) than summer and wintertime (R was about 0.38 and 0.56). These results were due to cloud contamination of summertime and the influence of snow/ice surface of wintertime, compared with those of other seasons. Therefore, the empirical multiple linear regression model used in this study showed that the AOT data retrieved from the satellite was important a dominant variable and we will need to use additional weather variables to improve the results of $PM_{2.5}$. Also, the result calculated for $PM_{2.5}$ using empirical multi linear regression model will be useful as a method to enable monitoring of atmospheric environment from satellite and ground meteorological data.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.40 no.2
• /
• pp.142-149
• /
• 1995

This study was conducted to investigate the relationships between yearly variation of climatic components and yearly variations of productivity in monoculture cotton. In addition, correlation coefficients among yield and yield components were estimated. The data of yield and yield components from the four varieties(Kinggus, Yongdang local. 113-4, 380) were collected from 1978 to 1992 in Mokpo area. The meteorological data gathered at the Mokpo Weather Station for the same period were used to find out the relationships between climatic components and productivity. Yearly variation of the amount of precipitation and number of stormy days in July are large with coefficients of the variations(C.V)84.89 and 97.05%, respectively, while yearly variation, of the average temperature, maximum temperature, minimum temperature from May to Sep. are relatively small. Seed cotton yield before frost in Sep. and Oct. very greatly with C.V. of 68.77, 78.52%, respectively. Number of boll bearing branches and lint percentage show more or less small in C.V. with 11.77 and 19.13%, respectively and flowering date and boll opening date show still less variation. Correlation coefficients between precipitation in May and number of boll bearing branches, duration of sunshine in July and number of bolls per plant, maximum temperature in July and total seed cotton before the frost in Sep., Oct., and Nov. evaporation in Aug. are positively sig-nificant at the 1% level. There are highly significantly positive correlated relationships among yield(total seed cotton) and yield components. Total seed cotton yield(Y) can be predicted by multiple regression equation with independent variables of climatic factors in July such as monthly averages of average temperature($X_1$), maximum temperature($X_2$) and minimum temperature($X_3$), monthly amount of precipitation ($X_4$), evaporation($X_5$), monthly average of relative humidity($X_6$), monthly hours with sunshine($X_7$) and number of rainy days($X_8$). The equation is estimatedas Y =-1080.8515 + 144.7133$X_1$+15.8722$X_2$ + 164.9367$X_3$ + 0.0802$X_4$ + 0.5932$X_5$ + 11.3373$X_6$ + 3.4683$X_7$- 9.0846$X_8$. Also, total seed cotton yield(Y) can be predicted by the same method with climatic components in Aug., Y =2835.2497 + 57.9134$X_1$ - 46.9055$X_2$ - 41.5886X$_3$ + 1.2559$X_5$ - 21.9687$X_6$ - 3.3763$X_7$- 4.1080$X_8$- 17.5586$X_9$. And the error between observed and theoretical yield were less with approached linear regression.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.34 no.3
• /
• pp.430-437
• /
• 2007

The probability table of Moyers and prediction equation of Tanaka and Johnston that have been the most frequently used, cannot produce accurate prediction when used in Korean because they are based on the Caucasian popularity of the Northern European race. The method of Moyers or Tanaka and Johnston predicts sizes of the unerupted canine and premolars on the basis of the sizes of mandibular incisors. However, some of the recent papers raise a question as to whether the mandibular incisors are the best combination to predict the sizes of the unerupted canine and premolars. The purpose of this study is to determine which sum or combination of sums of permanent tooth widths present the best prediction for the unerupted canine and premolars in a Korean sample, to calculate a specific linear regression equation for this population, and to evaluate the clinical significance. A new linear regression equation was calculated based on the data of 178 Korean young adults(70 women, 108 men, mean age 21.63 years) with complete permanent dentitions. Fifty three more children(28 girls, 25 boys, mean age 14.22 years) were used as a validation sample for the application of the multiple linear regression equation. The conclusions were as follows: 1. The combination of the sums of permanent upper central incisors, lower lateral incisors and upper first molars was the best predictor for the unerupted canine and premolars in this sample($r=0.65{\sim}0.80$). 2. The multiple linear regression equation was calculated including sex and arch as additional predictor variables. male, upper: $Y\;=\;0.332{\times}X_0\;+\;6.195$ male, lower: $Y\;=\;0.332{\times}X_0\;+\;5.269$ female, upper: $Y\;=\;0.332{\times}X_0\;+\;5.929$ female, lower: $Y\;=\;0.332{\times}X_0\;+\;5.003$. The determination coefficient of the equation was 64% and a standard error of the estimate was 0.71mm. 3. In about 97% of the validation sample, the estimation of the tooth width sums of unerupted canine and premolars using the new multiple linear regression equation was smaller than 1mm compaired with the actual values.