• Progress in Medical Physics
• /
• v.21 no.4
• /
• pp.332-339
• /
• 2010

We aimed to setup an adaptive radiation therapy platform using cone-beam CT (CBCT) and multileaf collimator (MLC) log data and also intended to analyze a trend of dose calculation errors during the procedure based on a phantom study. We took CT and CBCT images of Catphan-600 (The Phantom Laboratory, USA) phantom, and made a simple step-and-shoot intensity-modulated radiation therapy (IMRT) plan based on the CT. Original plan doses were recalculated based on the CT ($CT_{plan}$) and the CBCT ($CBCT_{plan}$). Delivered monitor unit weights and leaves-positions during beam delivery for each MLC segment were extracted from the MLC log data then we reconstructed delivered doses based on the CT ($CT_{recon}$) and CBCT ($CBCT_{recon}$) respectively using the extracted information. Dose calculation errors were evaluated by two-dimensional dose discrepancies ($CT_{plan}$ was the benchmark), gamma index and dose-volume histograms (DVHs). From the dose differences and DVHs, it was estimated that the delivered dose was slightly greater than the planned dose; however, it was insignificant. Gamma index result showed that dose calculation error on CBCT using planned or reconstructed data were relatively greater than CT based calculation. In addition, there were significant discrepancies on the edge of each beam while those were less than errors due to inconsistency of CT and CBCT. $CBCT_{recon}$ showed coupled effects of above two kinds of errors; however, total error was decreased even though overall uncertainty for the evaluation of delivered dose on the CBCT was increased. Therefore, it is necessary to evaluate dose calculation errors separately as a setup error, dose calculation error due to CBCT image quality and reconstructed dose error which is actually what we want to know.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.42 no.6
• /
• pp.21-30
• /
• 2014

The purpose of this study was to evaluate human thermal comfort in summer by the type of greenery in parks and to explore planning solutions to supply a comfortable thermal environment in parks. The research was conducted in three different land cover types: a park with multi-wide-canopied trees(WOODLAND), park with grass(LAWN) and park with pavement(PAV) as reference sites in Hamyang-Gun SangrimPark. Field measurements of air temperature, relative humidity and wind velocity, short-wave and long-wave radiation from six directions(east, west, north, south, upward and downward) were carried out in the summer of 2014(August 21-23 and 29-30). Mean Radiant Temperature($T_{mrt}$) absorbed by a human-biometeorological reference person was estimated from integral radiation and the calculation of angular factors. The thermal comfort index PET was calculated by Rayman software, UTCI, OUT_SET$^*$ were calculated using the UTCI Calculator and the Thermal Comfort Calculator of Richard DeDear. The results showed that the WOODLAND has the maximum cooling effect during daytime, reduced air temperatures/$T_{mrt}$ by up to $5.9^{\circ}C/35^{\circ}C$ compared to PAV and lowered heat stress values despite increasing relative humidity values and decreasing wind velocity. While the LAWN had very slight cooling effects during daytime, reduced air temperatures/$T_{mrt}$ by up to $0.9^{\circ}C/3^{\circ}C$ compared to PAV, the improvement effects of the thermal comfort index was very slight. However, during nighttime the microclimatic and radiant conditions of WOODLAND, LAWN, and PAV were similar owing to the absence of solar radiation, reduction of wind velocity and an increase in relative humidity. Because the shading and evapotranspiration effects of the WOODLAND were much greater than the evapotranspiration effects of the LAWN, it can be said that the solutions for supplying comfortable thermal environment in parks are to amplify the green volumes rather than green areas. This study was undertaken to evaluate the human thermal comfort in summer of WOODLAND/LAWN parks and to determine the improvement effects of thermal comfort index. These results can contribute to the provision better thermal comfort for park users during park planning.

• Journal of Intelligence and Information Systems
• /
• v.22 no.2
• /
• pp.127-142
• /
• 2016

Deep learning model is a kind of neural networks that allows multiple hidden layers. There are various deep learning architectures such as convolutional neural networks, deep belief networks and recurrent neural networks. Those have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. Among those architectures, convolutional neural networks and recurrent neural networks are classified as the supervised learning model. And in recent years, those supervised learning models have gained more popularity than unsupervised learning models such as deep belief networks, because supervised learning models have shown fashionable applications in such fields mentioned above. Deep learning models can be trained with backpropagation algorithm. Backpropagation is an abbreviation for "backward propagation of errors" and a common method of training artificial neural networks used in conjunction with an optimization method such as gradient descent. The method calculates the gradient of an error function with respect to all the weights in the network. The gradient is fed to the optimization method which in turn uses it to update the weights, in an attempt to minimize the error function. Convolutional neural networks use a special architecture which is particularly well-adapted to classify images. Using this architecture makes convolutional networks fast to train. This, in turn, helps us train deep, muti-layer networks, which are very good at classifying images. These days, deep convolutional networks are used in most neural networks for image recognition. Convolutional neural networks use three basic ideas: local receptive fields, shared weights, and pooling. By local receptive fields, we mean that each neuron in the first(or any) hidden layer will be connected to a small region of the input(or previous layer's) neurons. Shared weights mean that we're going to use the same weights and bias for each of the local receptive field. This means that all the neurons in the hidden layer detect exactly the same feature, just at different locations in the input image. In addition to the convolutional layers just described, convolutional neural networks also contain pooling layers. Pooling layers are usually used immediately after convolutional layers. What the pooling layers do is to simplify the information in the output from the convolutional layer. Recent convolutional network architectures have 10 to 20 hidden layers and billions of connections between units. Training deep learning networks has taken weeks several years ago, but thanks to progress in GPU and algorithm enhancement, training time has reduced to several hours. Neural networks with time-varying behavior are known as recurrent neural networks or RNNs. A recurrent neural network is a class of artificial neural network where connections between units form a directed cycle. This creates an internal state of the network which allows it to exhibit dynamic temporal behavior. Unlike feedforward neural networks, RNNs can use their internal memory to process arbitrary sequences of inputs. Early RNN models turned out to be very difficult to train, harder even than deep feedforward networks. The reason is the unstable gradient problem such as vanishing gradient and exploding gradient. The gradient can get smaller and smaller as it is propagated back through layers. This makes learning in early layers extremely slow. The problem actually gets worse in RNNs, since gradients aren't just propagated backward through layers, they're propagated backward through time. If the network runs for a long time, that can make the gradient extremely unstable and hard to learn from. It has been possible to incorporate an idea known as long short-term memory units (LSTMs) into RNNs. LSTMs make it much easier to get good results when training RNNs, and many recent papers make use of LSTMs or related ideas.

• Korean Journal of Heritage: History & Science
• /
• v.53 no.4
• /
• pp.78-99
• /
• 2020

The Buddha triad and 16 Arhat statues carved on the rock surface at Seongbulsa temple is the only domestic remaining example of all 16 Arhats, so its academic value is very high. However, it is severely damaged and so required a stability evaluation through study of digital documentation and precise diagnosis for the purpose of comprehensive conservation. This process established that the Buddha statues were of similar scale, while the Arhats showed a wide variety of sizes, and the two kith and kin in the volume were larger than the Arhats. It was estimated that the statues of food for Buddha are similar to the Arhat statues, and most of the statues are well-formed. The rock used to carve the Buddha statues is banded gneiss with distinct foliation, alternating between white bands of quartz and feldspar and black bands composed of biotite. The Buddha statues have been damaged by physical weathering, discoloration, and biological contamination. In damage evaluations, joint (3.6 crack index), peeling (5.2%), exfoliation (1.7%), and falling off (0.1%) were observed on the rock surface of the Buddha statues. In particular, due to severe biological weathering, stage 9 and 10 biological coverage of the rock surface accounted for 57.5% of the total area, and stages 5 to 8 also accounted for a high share at 22.3%. The discoloration factors were shown to be dark brown and white with Fe, Ca, and S, and a large amount of C detected in the blackened contaminants, and the damage weight high in all areas. Discontinuities in different directions were identified in the rock surface. Analysis of potential rock failure types indicated that there is a possibility of plane and toppling failure, but wedge failure is unlikely to occur. The mean ultrasonic velocity of the main rock surface was 2,463m/sec, the lower part of the left side with a large number of joints was relatively low, and the highly weathered (HW) type to the completely weathered (CW) type concentrated distribution, showing weak properties. For the Buddha statues, conservation treatment is required for about 14.9% of micro cracks and 58.9% of exfoliation cracks. In addition, in order to improve the conservation environment of the Buddha statues, maintenance of drainage and ground preparations for the rock surface gradient and plants are necessary, and protection facilities should be reviewed for long-term conservation and management purposes.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.18 no.3
• /
• pp.17-32
• /
• 2023

In recent years, the need for social ventures that aim to grow while solving social problems through the efficiency and effectiveness of commercial organizations in the market has increased, while there is a limit to how much the government and the public can do to solve social problems. Against this background, the number of social venture startups is increasing in the domestic startup ecosystem, and interest in impact investors, which are investors in social ventures, is also increasing. Therefore, this research utilized judgment analysis technology to objectively analyze the validity and weight of judgment information based on the cognitive process and decision-making environment in the investment decision-making of impact investors. We proceeded with the research by constructing three classifications; first, investment priorities at the initial investment stage for financial benefit and return on investment as an investor, second, the political skills of the entrepreneurs (teams) for the social impact and ripple power, and social venture coexistence and solidarity, third, the social mission of a social venture that meets the purpose of an impact investment fund. As a result of this research, first of all, the investment decision-making priorities of impact investors are the expertise of the entrepreneur (team), the potential rate of return when the entrepreneur (team) succeeds, and the social mission of the entrepreneur (team). Second, impact investors do not have a uniform understanding of the investment decision-making factors, and the factors that determine investment decisions are different, and there are differences in the degree of the weighting. Third, among the various investment decision-making factors of impact investment, "entrepreneur's (team's) networking ability", "entrepreneur's (team's) social insight", "entrepreneur's (team's) interpersonal influence" was relatively lower than the other four factors. The practical contribution through this research is to help social ventures understand the investment determinant factors of impact investors in the process of financing, and impact investors can be expected to improve the quality of investment decision-making by referring to the judgment cases and analysis of impact investors. The academic contribution is that it empirically investigated the investment priorities and weighting differences of impact investors.

• Journal of Nutrition and Health
• /
• v.49 no.5
• /
• pp.347-357
• /
• 2016

Purpose: This study investigated bone mineral density and nutritional state according to consumption of milk in Korean postmenopausal women who drink coffee. Methods: Using the 2008~2009 Korean National Health & Nutrition Examination Survey data, a total of 1,373 postmenopausal females aged 50 yrs and over were analyzed after excluding those with diseases related to bone health. According to coffee and/or milk consumption, subjects were divided into four groups: coffee only, both coffee & milk, milk only, and none of the above. All data were processed after application of weighted values and adjustment of age, body mass index, physical activity, drinking, and smoking using a general linear model. For analysis of nutrient intake and bone density, data were additionally adjusted by total energy and calcium intake. Results: The coffee & milk group had more subjects younger than 65 yrs and higher education, urban residents, and higher income than any other group. The coffee only group showed somewhat similar characteristics as the none of the above group, which showed the highest percentage of subjects older than 65 and in a lower education and socio-economic state. Body weight, height, body mass index, and lean mass were the highest in coffee & milk group and lowest in the none of the above group. On the other hand, the milk only group showed the lowest values for body mass index and waist circumference, whereas percent body fat did not show any difference among the groups. The coffee and milk group showed the highest bone mineral density in the total femur and lumbar spine as well as the highest nutritional state and most food group intakes, followed by the milk only group, coffee only group, and none of the above group. In the assessment of osteoporosis based on T-score of bone mineral density, although not significant, the coffee and milk group and milk only group, which showed a better nutritional state, included more subjects with a normal bone density, whereas the none of the above group included more subjects with osteoporosis than any other group. Conclusion: Bone mineral density in postmenopausal women might not be affected by coffee drinking if their diets are accompanied by balanced food and nutrient intake including milk.

• Journal of Intelligence and Information Systems
• /
• v.27 no.4
• /
• pp.73-95
• /
• 2021

This study uses Node2vec graph embedding method and Light GBM link prediction to explore undeveloped export candidate countries in Korea's food and beverage industry. Node2vec is the method that improves the limit of the structural equivalence representation of the network, which is known to be relatively weak compared to the existing link prediction method based on the number of common neighbors of the network. Therefore, the method is known to show excellent performance in both community detection and structural equivalence of the network. The vector value obtained by embedding the network in this way operates under the condition of a constant length from an arbitrarily designated starting point node. Therefore, it has the advantage that it is easy to apply the sequence of nodes as an input value to the model for downstream tasks such as Logistic Regression, Support Vector Machine, and Random Forest. Based on these features of the Node2vec graph embedding method, this study applied the above method to the international trade information of the Korean food and beverage industry. Through this, we intend to contribute to creating the effect of extensive margin diversification in Korea in the global value chain relationship of the industry. The optimal predictive model derived from the results of this study recorded a precision of 0.95 and a recall of 0.79, and an F1 score of 0.86, showing excellent performance. This performance was shown to be superior to that of the binary classifier based on Logistic Regression set as the baseline model. In the baseline model, a precision of 0.95 and a recall of 0.73 were recorded, and an F1 score of 0.83 was recorded. In addition, the light GBM-based optimal prediction model derived from this study showed superior performance than the link prediction model of previous studies, which is set as a benchmarking model in this study. The predictive model of the previous study recorded only a recall rate of 0.75, but the proposed model of this study showed better performance which recall rate is 0.79. The difference in the performance of the prediction results between benchmarking model and this study model is due to the model learning strategy. In this study, groups were classified by the trade value scale, and prediction models were trained differently for these groups. Specific methods are (1) a method of randomly masking and learning a model for all trades without setting specific conditions for trade value, (2) arbitrarily masking a part of the trades with an average trade value or higher and using the model method, and (3) a method of arbitrarily masking some of the trades with the top 25% or higher trade value and learning the model. As a result of the experiment, it was confirmed that the performance of the model trained by randomly masking some of the trades with the above-average trade value in this method was the best and appeared stably. It was found that most of the results of potential export candidates for Korea derived through the above model appeared appropriate through additional investigation. Combining the above, this study could suggest the practical utility of the link prediction method applying Node2vec and Light GBM. In addition, useful implications could be derived for weight update strategies that can perform better link prediction while training the model. On the other hand, this study also has policy utility because it is applied to trade transactions that have not been performed much in the research related to link prediction based on graph embedding. The results of this study support a rapid response to changes in the global value chain such as the recent US-China trade conflict or Japan's export regulations, and I think that it has sufficient usefulness as a tool for policy decision-making.