• Journal of Preventive Medicine and Public Health
• /
• v.15 no.1
• /
• pp.187-196
• /
• 1982

The followings are the results of the study on industrial accidents occurred at 12 factories manufacturing metal products during the period of 2 years from January 1980 to December 1981 in the area of Yong-Dung-Po in Seoul. The results of the study are as follows: 1. The incidence rate of industrial injuries was 45.7 per 1,000 workers of the sample group and the rate of male (54.0) was three times higher than that of female (17.5). 2. In age groups, the highest rate was observed in the group of under 19 years old with 83.5, while the lowest in the group of 40s. 3. It was found that those who had short term of work experience produced a higher rate of injuries, particularly, the group of workers with less than 1 year of experience showed the highest rate of it as 48.1%. 4. In working time, the highest incidence rate occurred 3 and 7 hours after the beginning of their working showing the rate of 6.0 and 6.1 per 1,000 workers, respectively. 5. The highest incidence rate was observed on Monday as 8.4 per 1,000 workers, and it was 18.3% in aspect of the days of a week. 6. In aspect of the months of a year, the highest incidence was observed on July 1,000 workers and the next was on March as 4.8. These figures account for 11.8% of total occurrence in respective month. as 5. 4 per and 10.5% 7. In causes of injuries, the accident caused by power driven machinery showed the highest rate with 37.5%, the second was due to handling without machinery with 17.2%, and the third was due to falling objects with 14.2%, and striking against objects with 10.2%, and so on. 8. By parts of the body affected, the most injuries 84.3% of them occurred on both upper and lower extremities with the rate of 58.8% for the former and 25.5% for the latter. Fingers were most frequently injured with a rate of 40.3%. Comparing the sites of extremities affected, rate of injuries on the right side was 55.0% and 45.0% on the left side. 9. In the nature of injury, laceration and open wound were the highest with 34. 0%, the next was fracture and dislocation with 31. 9%, and sprain was the third with 8.1%. 10. On the duration of treatment, it lasted less than one month in 68.9% of the injured cases, of which 14.5% of the cases were recovered within 2 weeks, and 54.4% of them were treated more than 2 weeks. And the duration of the treatment tended to be prolonged in larger industries. 11. The ratio of insured accidents to uninsured accidents was 1 to 4.7.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.15 no.4
• /
• pp.193-216
• /
• 2020

Due to the recent mid- to long-term slump and falling growth rates in the global economy, interest in organizational structures that create new products or services as a new alternative to survive and develop in an opaque environment both internally and externally, and enhance organizational sustainability through changes in production methods and business innovation is increasing day by day. In this atmosphere, we agree that the growth of small and medium-sized venture companies has a significant impact on the national economy, and various efforts are being made to enhance the technological innovation capabilities of the members so that these small and medium-sized venture companies can enhance and sustain their performance. The purpose of this study is also to investigate how the technological innovation capabilities of small and medium-sized venture companies correlate with the performance of knowledge management and to analyze the role of network capabilities to organize the strategic activities of enterprise to obtain the resources and organizational capabilities to be used for value creation from external networks. In other words, research was conducted on the impact of technological innovation capabilities of small and medium venture companies on knowledge management performance by using network capabilities as parameters. Therefore, in this study, we would like to verify the hypothesis that innovation capabilities will have a positive impact on knowledge management performance by using network capabilities of small and medium venture companies. Economic activities based on technological innovation capabilities should respond quickly to new changes in an environment where uncertainty has increased, and lead to macro-economic growth and development as well as overcoming long-term economic downturns so that they can become the nation's new growth engine as well as sustainable growth and survival of the organization. In addition, this study was conducted by setting the most important knowledge management performance within the organization as a dependent variable. As a result, R&D and learning capabilities among technological innovation capabilities have no impact on financial performance. In contrast, it was shown that corporate innovation activities have a positive impact on both financial and non-financial performance. The fact that non-financial factors such as quality and productivity improvement are identified in the management of small and medium-sized venture companies utilizing their technological innovation capabilities is contrary to a number of studies by those corporate innovation activities affect financial performance during prior research. The reason for this result is that research companies have been out of start-up companies for more than seven years, but sales are less than 10 billion won, and unlike start-up companies, R&D and learning capabilities have more positive effects on intangible non-financial performance than financial performance. Corporate innovation activities have been shown to have a positive (+) impact on both financial and non-financial performance, while R&D and learning capabilities have a positive (+) impact on financial performance by parameters of network capability. Corporate innovation activities have been shown to have no impact on both financial and non-financial performance, and R&D and learning capabilities have no impact on non-financial performance. It could be seen that the parameter effects of network competency are limited to when R&D and learning competencies are derived from quantitative financial performance. It could be seen that the parameter effects of network competency are limited to when R&D and learning competencies are derived from quantitative financial performance.

• The Journal of the Petrological Society of Korea
• /
• v.12 no.2
• /
• pp.79-99
• /
• 2003

Muscovite and biotite from 52 metasediments and 5 granites in the Hwasan area, the southwest of the Okcheon metamorphic belt and the Miwon-Jeungpyeong area, central Okcheon metamorphic belt were dated by the K-Ar and $^{40}$ Ar/$^{39}$ Ar methods. Muscovite and biotite ages from metapelitic and psammitic rocks (metasediments) of the Boeun and Pibanryeong units in the Hwasan area are concentrated in the mid-Jurassic (149-180 Ma). K-Ar and $^{40}$ Ar/$^{39}$ Ar ages for metapelitic and psammitic rocks of the Boeun and Pibanryeong units in the Miwon-Jeungpyeong area show complicated age distribution. Muscovite and biotite ages are classified by three groups, 142-194 Ma, 216-234 Ma, and 241-277 Ma. Younger (Cretaceous) ages occur only in metasediments close to Cretaceous granitic rocks in the southeastern region and the older ages of 216-277 Ma are restricted to the middle Part of the Jeungpyeong area. Most ages in the other area of the central Okcheon metamorphic belt fall between 142-194 Ma (Jurassic). K-Ar and $^{40}$ Ar/$^{39}$ Ar ages for granite from the northern part in the both the southwest and central Okcheon metamorphic belt also gave middle Jurassic ages (156-168 Ma). The similar ages from both metasediments and granites in the study areas indicate simultaneous cooling of both rocks to 300-350$^{\circ}C$ during the middle Jurassic. The state of graphitization of carbonaceous material of all metasediments in the study areas Indicates fully ordered graphite falling within a small range, from 3.353 to 3.359 ${\AA}$, which indicate amphibolite facies regional metamorphism. In the southern sector of the Boeun unit from the Hwasan area, metamorphic grade indicated by mineral paragenesis during regional intermediate-P/T metamorphism is greenschist facies. Whereas, the $d_{002}$ values for carbonaceous materials in the same sector show fully ordered graphite (ca. 500$^{\circ}C$) indicating amphibolite facies. This result with the concentration of mica ages of metasediments into the middle Jurassic, the presence of low-P/T thermal metamorphic zone (>500$^{\circ}C$) in the metasediments close to the Jurassic granite and the regional intrusion of Jurassic granites and their middle Jurassic intrusion and cooling ages may indicate the low-P/T regional thermal event during the early(\ulcorner)-middle Jurassic after main intermediate-P/T metamorphism which formed main mineral assemblage regionally in the study area. The regional thermal event failed, however, to reset the mineral assemblage of regional intermediate-P/T metamorphism except for narrow aureole (1-2 km) around Jurassic granite because e duration of thermal effect was relatively short by repid cooling of the Jurassic granite. In the middle part of the Jeungpyeong area, central Ogcheon metamorphic belt, muscovite and biotite K-Ar ages from 5 samples are 263-277 Ma and 241-249 Ma, respectively. An intermediate-P/T metamorphism is currently accepted to have occurred between 280 and 300 Ma. Therefore, the muscovite and biotite ages can be interpreted as cooling ages after Ml metamorphism indicating rapid cooling to ca 350$^{\circ}C$ between 280-300 Ma and 263-271 Ma, and biotite ages indicate slower cooling to ca. 300$^{\circ}C$ between 263-277 Ma and 241-249 Ma. However, more detail study is needed to confirm why the Permian to Triassic ages occur only in the middle Part of the Jeungpyeong area.a.

• Journal of Environmental Policy
• /
• v.13 no.4
• /
• pp.3-49
• /
• 2014

Growing interest of stakeholders on corporate responsibilities for environment and tightening environmental regulations are highlighting the importance of environmental management more than ever. However, companies' awareness of the importance of environment is still falling behind, and related academic works have not shown consistent conclusions on the relationship between environmental performance and economic performance. One of the reasons is different ways of measuring these two performances. The evaluation scope of economic performance is relatively narrow and the performance can be measured by a unified unit such as price, while the scope of environmental performance is diverse and a wide range of units are used for measuring environmental performances instead of using a single unified unit. Therefore, the results of works can be different depending on the performance indicators selected. In order to resolve this problem, generalized and standardized performance indicators should be developed. In particular, the performance indicators should be able to cover the concepts of both environmental and economic performances because the recent idea of environmental management has expanded to encompass the concept of sustainability. Another reason is that most of the current researches tend to focus on the motive of environmental investments and environmental performance, and do not offer a guideline for an effective implementation strategy for environmental management. For example, a process improvement strategy or a market discrimination strategy can be deployed through comparing the environment competitiveness among the companies in the same or similar industries, so that a virtuous cyclical relationship between environmental and economic performances can be secured. A novel method for measuring eco-efficiency by utilizing Data Envelopment Analysis (DEA), which is able to combine multiple environmental and economic performances, is proposed in this report. Based on the eco-efficiencies, the environmental competitiveness is analyzed and the optimal combination of inputs and outputs are recommended for improving the eco-efficiencies of inefficient firms. Furthermore, the panel analysis is applied to the causal relationship between eco-efficiency and economic performance, and the pooled regression model is used to investigate the relationship between eco-efficiency and economic performance. The four-year eco-efficiencies between 2010 and 2013 of 23 companies are obtained from the DEA analysis; a comparison of efficiencies among 23 companies is carried out in terms of technical efficiency(TE), pure technical efficiency(PTE) and scale efficiency(SE), and then a set of recommendations for optimal combination of inputs and outputs are suggested for the inefficient companies. Furthermore, the experimental results with the panel analysis have demonstrated the causality from eco-efficiency to economic performance. The results of the pooled regression have shown that eco-efficiency positively affect financial perform ances(ROA and ROS) of the companies, as well as firm values(Tobin Q, stock price, and stock returns). This report proposes a novel approach for generating standardized performance indicators obtained from multiple environmental and economic performances, so that it is able to enhance the generality of relevant researches and provide a deep insight into the sustainability of environmental management. Furthermore, using efficiency indicators obtained from the DEA model, the cause of change in eco-efficiency can be investigated and an effective strategy for environmental management can be suggested. Finally, this report can be a motive for environmental management by providing empirical evidence that environmental investments can improve economic performance.

• Journal of Intelligence and Information Systems
• /
• v.23 no.4
• /
• pp.147-168
• /
• 2017

There have been many studies on accurate stock market forecasting in academia for a long time, and now there are also various forecasting models using various techniques. Recently, many attempts have been made to predict the stock index using various machine learning methods including Deep Learning. Although the fundamental analysis and the technical analysis method are used for the analysis of the traditional stock investment transaction, the technical analysis method is more useful for the application of the short-term transaction prediction or statistical and mathematical techniques. Most of the studies that have been conducted using these technical indicators have studied the model of predicting stock prices by binary classification - rising or falling - of stock market fluctuations in the future market (usually next trading day). However, it is also true that this binary classification has many unfavorable aspects in predicting trends, identifying trading signals, or signaling portfolio rebalancing. In this study, we try to predict the stock index by expanding the stock index trend (upward trend, boxed, downward trend) to the multiple classification system in the existing binary index method. In order to solve this multi-classification problem, a technique such as Multinomial Logistic Regression Analysis (MLOGIT), Multiple Discriminant Analysis (MDA) or Artificial Neural Networks (ANN) we propose an optimization model using Genetic Algorithm as a wrapper for improving the performance of this model using Multi-classification Support Vector Machines (MSVM), which has proved to be superior in prediction performance. In particular, the proposed model named GA-MSVM is designed to maximize model performance by optimizing not only the kernel function parameters of MSVM, but also the optimal selection of input variables (feature selection) as well as instance selection. In order to verify the performance of the proposed model, we applied the proposed method to the real data. The results show that the proposed method is more effective than the conventional multivariate SVM, which has been known to show the best prediction performance up to now, as well as existing artificial intelligence / data mining techniques such as MDA, MLOGIT, CBR, and it is confirmed that the prediction performance is better than this. Especially, it has been confirmed that the 'instance selection' plays a very important role in predicting the stock index trend, and it is confirmed that the improvement effect of the model is more important than other factors. To verify the usefulness of GA-MSVM, we applied it to Korea's real KOSPI200 stock index trend forecast. Our research is primarily aimed at predicting trend segments to capture signal acquisition or short-term trend transition points. The experimental data set includes technical indicators such as the price and volatility index (2004 ~ 2017) and macroeconomic data (interest rate, exchange rate, S&P 500, etc.) of KOSPI200 stock index in Korea. Using a variety of statistical methods including one-way ANOVA and stepwise MDA, 15 indicators were selected as candidate independent variables. The dependent variable, trend classification, was classified into three states: 1 (upward trend), 0 (boxed), and -1 (downward trend). 70% of the total data for each class was used for training and the remaining 30% was used for verifying. To verify the performance of the proposed model, several comparative model experiments such as MDA, MLOGIT, CBR, ANN and MSVM were conducted. MSVM has adopted the One-Against-One (OAO) approach, which is known as the most accurate approach among the various MSVM approaches. Although there are some limitations, the final experimental results demonstrate that the proposed model, GA-MSVM, performs at a significantly higher level than all comparative models.

• Information Systems Review
• /
• v.16 no.3
• /
• pp.161-177
• /
• 2014

Corporate credit rating assessment consists of complicated processes in which various factors describing a company are taken into consideration. Such assessment is known to be very expensive since domain experts should be employed to assess the ratings. As a result, the data-driven corporate credit rating prediction using statistical and artificial intelligence (AI) techniques has received considerable attention from researchers and practitioners. In particular, statistical methods such as multiple discriminant analysis (MDA) and multinomial logistic regression analysis (MLOGIT), and AI methods including case-based reasoning (CBR), artificial neural network (ANN), and multiclass support vector machine (MSVM) have been applied to corporate credit rating.2) Among them, MSVM has recently become popular because of its robustness and high prediction accuracy. In this study, we propose a novel optimized MSVM model, and appy it to corporate credit rating prediction in order to enhance the accuracy. Our model, named 'GAMSVM (Genetic Algorithm-optimized Multiclass Support Vector Machine),' is designed to simultaneously optimize the kernel parameters and the feature subset selection. Prior studies like Lorena and de Carvalho (2008), and Chatterjee (2013) show that proper kernel parameters may improve the performance of MSVMs. Also, the results from the studies such as Shieh and Yang (2008) and Chatterjee (2013) imply that appropriate feature selection may lead to higher prediction accuracy. Based on these prior studies, we propose to apply GAMSVM to corporate credit rating prediction. As a tool for optimizing the kernel parameters and the feature subset selection, we suggest genetic algorithm (GA). GA is known as an efficient and effective search method that attempts to simulate the biological evolution phenomenon. By applying genetic operations such as selection, crossover, and mutation, it is designed to gradually improve the search results. Especially, mutation operator prevents GA from falling into the local optima, thus we can find the globally optimal or near-optimal solution using it. GA has popularly been applied to search optimal parameters or feature subset selections of AI techniques including MSVM. With these reasons, we also adopt GA as an optimization tool. To empirically validate the usefulness of GAMSVM, we applied it to a real-world case of credit rating in Korea. Our application is in bond rating, which is the most frequently studied area of credit rating for specific debt issues or other financial obligations. The experimental dataset was collected from a large credit rating company in South Korea. It contained 39 financial ratios of 1,295 companies in the manufacturing industry, and their credit ratings. Using various statistical methods including the one-way ANOVA and the stepwise MDA, we selected 14 financial ratios as the candidate independent variables. The dependent variable, i.e. credit rating, was labeled as four classes: 1(A1); 2(A2); 3(A3); 4(B and C). 80 percent of total data for each class was used for training, and remaining 20 percent was used for validation. And, to overcome small sample size, we applied five-fold cross validation to our dataset. In order to examine the competitiveness of the proposed model, we also experimented several comparative models including MDA, MLOGIT, CBR, ANN and MSVM. In case of MSVM, we adopted One-Against-One (OAO) and DAGSVM (Directed Acyclic Graph SVM) approaches because they are known to be the most accurate approaches among various MSVM approaches. GAMSVM was implemented using LIBSVM-an open-source software, and Evolver 5.5-a commercial software enables GA. Other comparative models were experimented using various statistical and AI packages such as SPSS for Windows, Neuroshell, and Microsoft Excel VBA (Visual Basic for Applications). Experimental results showed that the proposed model-GAMSVM-outperformed all the competitive models. In addition, the model was found to use less independent variables, but to show higher accuracy. In our experiments, five variables such as X7 (total debt), X9 (sales per employee), X13 (years after founded), X15 (accumulated earning to total asset), and X39 (the index related to the cash flows from operating activity) were found to be the most important factors in predicting the corporate credit ratings. However, the values of the finally selected kernel parameters were found to be almost same among the data subsets. To examine whether the predictive performance of GAMSVM was significantly greater than those of other models, we used the McNemar test. As a result, we found that GAMSVM was better than MDA, MLOGIT, CBR, and ANN at the 1% significance level, and better than OAO and DAGSVM at the 5% significance level.

• Journal of Korean Society of Forest Science
• /
• v.88 no.4
• /
• pp.510-522
• /
• 1999

The air pollutants ; $SO_2$, $SO{_4}^{-2}$, $NO{_3}^-$, $Cl^-$ are absorbed into soils through falling with dusts and rain from the atmosphere. The sources of heavy metal contaminants in the environments are agricultural and horticultural materials, sewage sludges, fossil fuel combustion, metallurgical industries, electronics and waste disposal etc.. The soils and hydrosphere can be polluted on the way of the circulation of these heavy metals. Studied pollutant anions are $SO{_4}^{-2}$, $NO{_3}^-$ and $Cl^-$ and heavy metals are Se, Mo, Zn, Cd, Pb, Mn, Cr, Co, V, As, Cu and Ni which are the elements to be concerned with the essentials for plants, with animal and human health. This study is with the aim of selecting the species of roadside trees and green space trees which have excellent absorption of air pollutants and heavy metals from the atmosphere and the soils in the urban area. Two areas are designated to carry out this study : urban area ; Kwangju city and rural area ; the yard of Forest Environment Institute of Chollanam-do, at Sanje-ri, Sampo-myum, Naju city, Chollanam-do (23km away from Kwangju). This study is carried out to understand the movement of anions and heavy metals from the soils to the trees in both areas, the absorption of anions and heavy metals from atmosphere into leaves and the amounts of anions and heavy metals in leaves and fine roots(< 1mm dia.) of roadside trees and green space trees in Kwangju and trees in the yard of Forest Environment Institute of Chollanam-do. The tree species selected for this study in both areas are Ginkgo biloba, Quercus acutissima, Cedrus deodara, Platanus occidentalis, Robinia pseudoacacia, Alnus japonica. Metasequoia glyptostroboides. Zekova serrata. Prunus serrulata var. spontanea, and Pinus densiflora. The results of the study are as follows : 1. $SO{_4}^{-2}$, $NO{_3}^-$ and $Cl^-$ concentrations are higher in the soils of the urban area than in those of the rural area, and $NO{_3}^-$ and $SO{_4}^{-2}$ are higher in the leaves than in the roots due to the absorption of the these pollutants through the stomata. 2. Ginkgo biloba, Robinia pseudoacacia. Zekova serrata, Quercus acutissima, and Platanus occidentalis can be adequated to the roadside trees and the environmental trees due to their good absorption of $NO{_3}^-$ and $SO{_4}^{-2}$. 3. Heavy metals in the soils of both areas are in the order of Mn > Zn > V > Cr > Pb > Ni > Cu > Mo> Cd, and in the leaves and roots of the trees in the both areas are in the order of Mn>Zn>Cr>Cu>V>Ni. Both orders are similar ones except V. There are more in the urban soils than in the rural soils in amount of Mn, Zn, Pb, V, Cu. 4. It is supposed that there is no antagonism between Mn and Zn in this study. 5. Se, Co and As are not detected in the soils, the leaves and the roots in both areas. Sn, Mo, Cd and Pb are also not detected in the leaves and roots in spite of considerable amount in the soils of both areas.

• Journal of dental hygiene science
• /
• v.4 no.1
• /
• pp.21-25
• /
• 2004

In modern times, children's trauma is increasing every year because of car accidents and life environment changes. There is a limit to prevent traumatic damage for oral cavity organization. The fundamental data of trauma treatment and prevention will be presented through the survey and analysis of traumatic teeth damage. I examined 113 patients from Oct. 4th, 2000 to Feb. 27th, 2004 at Dept. of Children's Dental Clinic, Kangnung National University. The results are as follows. (1) The trauma frequency of male subjects is higher than that of female at a rate of 2.05:1. The average age is 5.27 for men and 5.27 for women. The highest percentage of trauma patients is among 2 year old children. It is 21.2%. (2) A patient survey was taken at a trauma treatment hospital. On the first day 34.4% of the patients had come to receive treatment of their first set of teeth. However, after a week, 38.8% of the patients had received treatment on their permanent teeth. (3) As a result of falling, 59% of patients needing treatment on their first set of teeth. 55.1% of patients is permanent teeth. As a result of bump against physical solid, 26.6% of patients is the first set of teeth and 26.5% of patients is permanent teeth. (4) Teeth damage happened at home. 42.1% were male. 35.1% were female. According to trauma, 59.4% of teeth damage happened at home. 28.6% of permanent teeth damage happened at school or kindergarten. (5) According to trauma, the number of teeth damaged was in the first set of teeth are as follows: 56.3%, one-31.3%, three or four-6.3% each. For permanent teeth: two-46.9%, one-28.6%, four over-16.3% and three-8.2%. Over four teeth is larger number for permanent teeth. (6) 56% of first set of teeth patients and 43.4% of permanent teeth patients were male. 56.8% of first set of teeth patients and 43.2% of permanent teeth were female. Trauma happened to both male and female frequently in the first set of teeth. (7) Most of the tooth damage which was in the first set of teeth and permanent teeth was done to the upper jaw. 75% of patients are the first set of teeth. 63.8% of patients are permanent teeth. Trauma is very high in the two mid teeth of the upper jaw. (8) According to trauma survey, 30.2% is from impulse. 28.0% is from crown fracture, 14.7% is from depression. 8.9% is from concussion. 7.1% is from full dislocation of a joint. 2.2% of patients are extrusion. 1.8% is from displacement. According to teeth damage trauma, 35.8% is pulse in the first set of teeth. The breaking of the crown of a tooth happened a lot in permanent teeth. (9) According to data, 43.2% of teeth damage in the first set of teeth goes without treatment. In permanent teeth, it is 38.9%. After treatment, 22.0% of first set of teeth treatment requires a dental pulp treatment. In permanent teeth, which is used for temporary acid etching resin restoration.

• Journal of Intelligence and Information Systems
• /
• v.19 no.4
• /
• pp.123-132
• /
• 2013

As the smartphones are equipped with various sensors such as the accelerometer, GPS, gravity sensor, gyros, ambient light sensor, proximity sensor, and so on, there have been many research works on making use of these sensors to create valuable applications. Human activity recognition is one such application that is motivated by various welfare applications such as the support for the elderly, measurement of calorie consumption, analysis of lifestyles, analysis of exercise patterns, and so on. One of the challenges faced when using the smartphone sensors for activity recognition is that the number of sensors used should be minimized to save the battery power. When the number of sensors used are restricted, it is difficult to realize a highly accurate activity recognizer or a classifier because it is hard to distinguish between subtly different activities relying on only limited information. The difficulty gets especially severe when the number of different activity classes to be distinguished is very large. In this paper, we show that a fairly accurate classifier can be built that can distinguish ten different activities by using only a single sensor data, i.e., the smartphone accelerometer data. The approach that we take to dealing with this ten-class problem is to use the ensemble of nested dichotomy (END) method that transforms a multi-class problem into multiple two-class problems. END builds a committee of binary classifiers in a nested fashion using a binary tree. At the root of the binary tree, the set of all the classes are split into two subsets of classes by using a binary classifier. At a child node of the tree, a subset of classes is again split into two smaller subsets by using another binary classifier. Continuing in this way, we can obtain a binary tree where each leaf node contains a single class. This binary tree can be viewed as a nested dichotomy that can make multi-class predictions. Depending on how a set of classes are split into two subsets at each node, the final tree that we obtain can be different. Since there can be some classes that are correlated, a particular tree may perform better than the others. However, we can hardly identify the best tree without deep domain knowledge. The END method copes with this problem by building multiple dichotomy trees randomly during learning, and then combining the predictions made by each tree during classification. The END method is generally known to perform well even when the base learner is unable to model complex decision boundaries As the base classifier at each node of the dichotomy, we have used another ensemble classifier called the random forest. A random forest is built by repeatedly generating a decision tree each time with a different random subset of features using a bootstrap sample. By combining bagging with random feature subset selection, a random forest enjoys the advantage of having more diverse ensemble members than a simple bagging. As an overall result, our ensemble of nested dichotomy can actually be seen as a committee of committees of decision trees that can deal with a multi-class problem with high accuracy. The ten classes of activities that we distinguish in this paper are 'Sitting', 'Standing', 'Walking', 'Running', 'Walking Uphill', 'Walking Downhill', 'Running Uphill', 'Running Downhill', 'Falling', and 'Hobbling'. The features used for classifying these activities include not only the magnitude of acceleration vector at each time point but also the maximum, the minimum, and the standard deviation of vector magnitude within a time window of the last 2 seconds, etc. For experiments to compare the performance of END with those of other methods, the accelerometer data has been collected at every 0.1 second for 2 minutes for each activity from 5 volunteers. Among these 5,900 ($=5{\times}(60{\times}2-2)/0.1$) data collected for each activity (the data for the first 2 seconds are trashed because they do not have time window data), 4,700 have been used for training and the rest for testing. Although 'Walking Uphill' is often confused with some other similar activities, END has been found to classify all of the ten activities with a fairly high accuracy of 98.4%. On the other hand, the accuracies achieved by a decision tree, a k-nearest neighbor, and a one-versus-rest support vector machine have been observed as 97.6%, 96.5%, and 97.6%, respectively.

• Journal of Sasang Constitutional Medicine
• /
• v.10 no.2
• /
• pp.351-429
• /
• 1998

The Purpose of Study 1. Inspection of clinical application on TCD to CVA 2. Objective Comparement and analysis about treatment effect of Western-Medicine, Korean Medicine, Cooperative consultation of Korean and Western medicice for CVA The Subject of Study We intended for the eighty six patient of CVA who had been treated in the Oriental Medical Hospital at Dong Eui Medical Center from 1997. 8. I to 1998. 7. 31 1. View of CT, MRI : the patient of Cb infarction 2. Attack Time : The patient who coming hospital falling ill within the early one week The method of study 1. Treat four group of Korean medicine, Constitution medicine, Western medicine, cooperative consultation of Korean medicine and Western medicine. 2. Application of TCD Check the result for three times, immediatly after the attack, two months later, four months later 3. Comparative analysis of each treatment effect by clinical symptoms and pathologic examination 4. The Judgement of the patient The Result From 8/1/1997 to 7/31/1998, We have the following result by clinical analysis intended for CVA 86 patients who had been treated in the Oriental Medical Hospital at Dong Eui Medical Center from 1997. 8. 1 to 1998. 7. 31 in 1. Analysis according to Age The first stage of thirties, forties, seventies is heavier than forties, fifties in improvement and Index of improvement of symptom 2. Analysis according to sex We have no special relation in an average of symptom and improvement, Index improvement 3. Analysis according to Family History We have the better result in first stage and improvement, index improvement when no family history. 4. Analysis according to Past History We have no special relation in past history like hypertension, DM, heart problem 5. Analysis devided two group, above group and under group on the basis of the average in first stage of all patient. We have the better result when the first stage is light, that the first score of barthel index and CNS is high. 6. Analysis of the effect of treatment about Korean medical treatment, Western medical treatment, cooperative treatment. In this study, the highest group of rate of treatment at four contrast groups (Korean medicine, Constitution medicine, Western medicine, cooperative treatment according to dyagnosis and range of treatment was the patient group of doing dyagnosis and method of treatment based on constitution medicine theory. This is that of doing demostation, A-Tx, po-herb-medication according to dyagnosis and treat method of constitution of Lee Je-ma In case of left, the case of dyagnosis any disease according to doctor view but, normal in TCDwas 22-beginning of attack, 20- two weeks later, 11 case-four weeks later in case of right, 15-beginning of attack, 12-two weeks later, 9 case four weeks later. So left vessel compares to right vessel is more interference, in fact more than a 1/2 of the patients of MCA disease can't do dyagnosis. In rate of imparement, the state of pacient improved but there isn't the improved case of result in TCD. 7. In TCD dyagnosis, between the case of inconsus the doctor view specially MCA in brain blood vessel is in large numbers and in total 86's patient, impossible case of dyagnosis according to interferiance of temporal is 21 case. 7. Result study about application of Kreaan medical treatment 1) The impossible patient of observation MCA blood vescular for interference temporal bone happened in large numbers. 2) There is the case having difference result to CT,MRI, MRA result. 3) Because individual difference is large, excluding to ananalogy of symptom. This is normal numerical value that has possibility of being checked as abnormal numerical value 4) there are a lot of cases that the speed of normal part is as similarly measured as that of abnormal part. It means that we cannot judge the disease by this measure 5) It is rare that this measure represent degree of improvement in patient's condition of disease. When we observe patient's condition become better, but we have no case that the result of TCD test better. 6) The result could be appear differently by the technique of the tester or by the experience of the tester 7) In the TCD test, abnormal symptoms is checked at 0 week, but at 2th week, normal symptoms is checked, again at 4th week abnormal is checked. According to the above result, CVA diagnosis is difficult only with TCD, as it appear in diagnosis error check which is suggested in the problem connected to project, for the aged persons who have the worst hardening of part of the cranium (1998. 5. 26 77 of 83 patients is 50s) there is a lot of cases that the measurement is impossible by TCD and the correction of measurement numerical value is decreased, as the age of cerebral infarction is high, TCD is inappropriate to diagnosis equipment through this study.