• Radiation Oncology Journal
• /
• 제24권4호
• /
• pp.294-299
• /
• 2006

$\underline{Purpose}$: Using cone beam CT, we can compare the position of the patients at the simulation and the treatment. In on-line image guided radiation therapy, one can utilize this compared data and correct the patient position before treatments. Using cone beam CT, we investigated the errors induced by setting up the patients when use only the markings on the patients' skin. $\underline{Materials\;and\;Methods}$: We obtained the data of three patients that received radiation therapy at the Department of Radiation Oncology in Chung-Ang University during August 2006 and October 2006. Just as normal radiation therapy, patients were aligned on the treatment couch after the simulation and treatment planning. Patients were aligned with lasers according to the marking on the skin that were marked at the simulation time and then cone beam CTs were obtained. Cone beam CTs were fused and compared with simulation CTs and the displacement vectors were calculated. Treatment couches were adjusted according to the displacement vector before treatments. After the treatment, positions were verified with kV X-ray (OBI system). $\underline{Results}$: In the case of head and neck patients, the average sizes of the setup error vectors, given by the cone beam CT, were 0.19 cm for the patient A and 0.18 cm for the patient B. The standard deviations were 0.15 cm and 0.21 cm, each. On the other hand, in the case of the pelvis patient, the average and the standard deviation were 0.37 cm and 0.1 cm. $\underline{Conclusion}$: Through the on-line IGRT using cone beam CT, we could correct the setup errors that could occur in the conventional radiotherapy. The importance of the on-line IGRT should be emphasized in the case of 3D conformal therapy and intensity-modulated radiotherapy, which have complex target shapes and steep dose gradients.

• Korean Journal of Remote Sensing
• /
• 제35권6_2호
• /
• pp.1117-1132
• /
• 2019

Forest covers 30% of the Earth's land area and plays an important role in global carbon flux through its ability to store much greater amounts of carbon than other terrestrial ecosystems. The Gross Primary Production (GPP) represents the productivity of forest ecosystems according to climate change and its effect on the phenology, health, and carbon cycle. In this study, we estimated the daily GPP for a forest ecosystem using remote-sensed data from Moderate Resolution Imaging Spectroradiometer (MODIS) and machine learning algorithms Support Vector Machine (SVM). MODIS products were employed to train the SVM model from 75% to 80% data of the total study period and validated using eddy covariance measurement (EC) data at the six flux tower sites. We also compare the GPP derived from EC and MODIS (MYD17). The MODIS products made use of two data sets: one for Processed MODIS that included calculated by combined products (e.g., Vapor Pressure Deficit), another one for Unprocessed MODIS that used MODIS products without any combined calculation. Statistical analyses, including Pearson correlation coefficient (R), mean squared error (MSE), and root mean square error (RMSE) were used to evaluate the outcomes of the model. In general, the SVM model trained by the Unprocessed MODIS (R = 0.77 - 0.94, p < 0.001) derived from the multi-sites outperformed those trained at a single-site (R = 0.75 - 0.95, p < 0.001). These results show better performance trained by the data including various events and suggest the possibility of using remote-sensed data without complex processes to estimate GPP such as non-stationary ecological processes.

• Journal of Korea Port Economic Association
• /
• 제36권4호
• /
• pp.17-30
• /
• 2020

In the shipping industry, it is essential to engage in the preemptive prediction of freight rate volatility through market monitoring. Considering that freight rates have already started to fall, the loss of shipping companies will soon be uncontrollable. Therefore, in this study, factors affecting the freight rates of bulk carriers, which have relatively large freight rate volatility as compared to container freight rates, were quantified and analyzed. In doing so, we intended to contribute to future shipping market monitoring. We performed an analysis using a vector error correction model and estimated the influence of six independent variables on the charter rates of bulk carriers by Handy Size, Supramax, Panamax, and Cape Size. The six independent variables included the bulk carrier fleet volume, iron ore traffic volume, ribo interest rate, bunker oil price, and Euro-Dollar exchange rate. The dependent variables were handy size (32,000 DWT) spot charter rates, Supramax 6 T/C average charter rates, Pana Max (75,000 DWT) spot charter, and Cape Size (170,000 DWT) spot charter. The study examined charter rates by size of bulk carriers, which was different from studies on existing specific types of ships or fares in oil tankers and chemical carriers other than bulk carriers. Findings revealed that influencing factors differed for each ship size. The Libo interest rate had a significant effect on all four ship types, and the iron ore traffic volume had a significant effect on three ship types. The Ribo rate showed a negative (-) relationship with Handy Size, Supramax, Panamax, and Cape Size. Iron ore traffic influenced three types of linearity, except for Panamax. The size of shipping companies differed depending on their characteristics. These findings are expected to contribute to the establishment of a management strategy for shipping companies by analyzing the factors influencing changes in the freight rates of charterers, which have a profound effect on the management performance of shipping companies.

• Korean Journal of Agricultural and Forest Meteorology
• /
• 제21권1호
• /
• pp.29-41
• /
• 2019

Terrestrial Gross Primary Production (GPP) is the largest global carbon flux, and forest ecosystems are important because of the ability to store much more significant amounts of carbon than other terrestrial ecosystems. There have been several attempts to estimate GPP using mechanism-based models. However, mechanism-based models including biological, chemical, and physical processes are limited due to a lack of flexibility in predicting non-stationary ecological processes, which are caused by a local and global change. Instead mechanism-free methods are strongly recommended to estimate nonlinear dynamics that occur in nature like GPP. Therefore, we used the mechanism-free machine learning techniques to estimate the daily GPP. In this study, support vector machine (SVM), random forest (RF) and artificial neural network (ANN) were used and compared with the traditional multiple linear regression model (LM). MODIS products and meteorological parameters from eddy covariance data were employed to train the machine learning and LM models from 2006 to 2013. GPP prediction models were compared with daily GPP from eddy covariance measurement in a deciduous forest in South Korea in 2014 and 2015. Statistical analysis including correlation coefficient (R), root mean square error (RMSE) and mean squared error (MSE) were used to evaluate the performance of models. In general, the models from machine-learning algorithms (R = 0.85 - 0.93, MSE = 1.00 - 2.05, p < 0.001) showed better performance than linear regression model (R = 0.82 - 0.92, MSE = 1.24 - 2.45, p < 0.001). These results provide insight into high predictability and the possibility of expansion through the use of the mechanism-free machine-learning models and remote sensing for predicting non-stationary ecological processes such as seasonal GPP.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• 제13권1호
• /
• pp.42-55
• /
• 2008

To evaluate the accuracy of currents measured by HF radar in the coastal sea off Keum River estuary, we compared the facing radial vectors of two HF radars, and HF radar-derived currents with in-situ measurement currents. Principal component analysis was used to extract regression line and RMS deviation in the comparison. When two facing radar's radial vectors at the mid-point of baseline are compared, RMS deviation is 4.4 cm/s in winter and 5.4 cm/s in summer. When GDOP(Geometric Dilution of Precision) effect is corrected from the RMS deviations that is analyzed from the comparison between HF radar-derived and current-metermeasured currents, the error of velocity combined by HF radar-derived current is less than 5.1 cm/s in the stations having moderate GDOP values. These two results obtained from different method suggest that the lower limit of HF radar-derived current's accuracy is 5.4 cm/s in our study area. As mentioned in previous researches, RMS deviations become large in the stations located near the islands and increase as a function of mean distance from the radar site due to decrease of signal-to-noise level and the intersect angle of radial vectors. We found that an uncertain error bound of HF radar-derived current can be produced from the separation process of RMS deviations using GDOP value if GDOP value for each component is very close and RMS deviations obtained from current component comparison are also close. When the current measured in the stations having moderate GDOP values is separated into tidal and subtidal current, characteristics of tidal current ellipses analyzed from HF radar-derived current show a good agreement with those from current-meter-measured current, and time variation of subtidal current showed a response reflecting physical process driven by wind and density field.

• Asia pacific journal of information systems
• /
• 제19권2호
• /
• 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.

• KDI Journal of Economic Policy
• /
• 제31권1호
• /
• pp.169-206
• /
• 2009

This study stems from a question, "How should we understand the pattern of the Korean economy after the 1990s?" Among various analytic methods applicable, this study chooses a Structural Vector Autoregression (SVAR) with long-run restrictions, identifies diverse impacts that gave rise to the current status of the Korean economy, and differentiates relative contributions of those impacts. To that end, SVAR is applied to four economic models; Blanchard and Quah (1989)'s 2-variable model, its 3-variable extensions, and the two other New Keynesian type linear models modified from Stock and Watson (2002). Especially, the latter two models are devised to reflect the recent transitions in the determination of foreign exchange rate (from a fixed rate regime to a flexible rate one) as well as the monetary policy rule (from aggregate targeting to inflation targeting). When organizing the assumed results in the form of impulse response and forecasting error variance decomposition, two common denominators are found as follows. First, changes in the rate of economic growth are mainly attributable to the impact on productivity, and such trend has grown strong since the 2000s, which indicates that Korea's economic growth since the 2000s has been closely associated with its potential growth rate. Second, the magnitude or consistency of impact responses tends to have subsided since the 2000s. Given Korea's high dependence on trade, it is possible that low interest rates, low inflation, steady growth, and the economic emergence of China as a world player have helped secure capital and demand for export and import, which therefore might reduced the impact of each sector on overall economic status. Despite the fact that a diverse mixture of models and impacts has been used for analysis, always two common findings are observed in the result. Therefore, it can be concluded that the decreased rate of economic growth of Korea since 2000 appears to be on the same track as the decrease in Korea's potential growth rate. The contents of this paper are constructed as follows: The second section observes the recent trend of the economic development of Korea and related Korean articles, which might help in clearly defining the scope and analytic methodology of this study. The third section provides an analysis model to be used in this study, which is Structural VAR as mentioned above. Variables used, estimation equations, and identification conditions of impacts are explained. The fourth section reports estimation results derived by the previously introduced model, and the fifth section concludes.

• Korean Journal of Remote Sensing
• /
• 제37권6_1호
• /
• pp.1739-1756
• /
• 2021

Atmospheric nitrogen dioxide (NO₂) is mainly caused by anthropogenic emissions. It contributes to the formation of secondary pollutants and ozone through chemical reactions, and adversely affects human health. Although ground stations to monitor NO₂ concentrations in real time are operated in Korea, they have a limitation that it is difficult to analyze the spatial distribution of NO₂ concentrations, especially over the areas with no stations. Therefore, this study conducted a comparative experiment of spatial interpolation of NO₂ concentrations based on two linear-regression methods(i.e., multi linear regression (MLR), and regression kriging (RK)), and two machine learning approaches (i.e., random forest (RF), and support vector regression (SVR)) for the year of 2020. Four approaches were compared using leave-one-out-cross validation (LOOCV). The daily LOOCV results showed that MLR, RK, and SVR produced the average daily index of agreement (IOA) of 0.57, which was higher than that of RF (0.50). The average daily normalized root mean square error of RK was 0.9483%, which was slightly lower than those of the other models. MLR, RK and SVR showed similar seasonal distribution patterns, and the dynamic range of the resultant NO₂ concentrations from these three models was similar while that from RF was relatively small. The multivariate linear regression approaches are expected to be a promising method for spatial interpolation of ground-level NO₂ concentrations and other parameters in urban areas.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• 제19권8호
• /
• pp.353-359
• /
• 2018

In this study, the performance between subjective refraction and open-field/closed view autorefraction was estimated. We measured the refractive error of early adults aged 18 to 20 years who did not have eye disease. The differences between measurements obtained by subjective refraction and open-field autorefraction for SE, J0, and J45 were $-0.13{\pm}0.53D$ (p=0.17), $+0.33{\pm}0.68D$ (p=0.01), and $+0.13{\pm}0.68D$ (p=0.26), respectively, with only J0 differing significantly. The differences between the measurements of subjective refraction and closed-view autorefraction for SE, J0, and J45 were $-0.30{\pm}0.42D$ (p=0.00), $+0.30{\pm}0.71D$ (p=0.02), and $-0.02{\pm}0.63D$ (p=0.88), respectively, with only SE and J0 differing significantly. The coefficient of accuracy for SE, J0, and J45 components of open-field and closed-view autorefraction were 1.04, 1.33, and 1.34 and 0.83, 1.40, and 1.24, respectively. It is possible to predict the refractive error, which is necessary when deciding on subjective refraction, by measuring the objective refraction of open-field/closed view autorefractors.

• Journal of Intelligence and Information Systems
• /
• 제18권2호
• /
• 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.