• Journal of Intelligence and Information Systems
• /
• v.16 no.4
• /
• pp.99-112
• /
• 2010

Ensemble learning is a method for improving the performance of classification and prediction algorithms. It is a method for finding a highly accurateclassifier on the training set by constructing and combining an ensemble of weak classifiers, each of which needs only to be moderately accurate on the training set. Ensemble learning has received considerable attention from machine learning and artificial intelligence fields because of its remarkable performance improvement and flexible integration with the traditional learning algorithms such as decision tree (DT), neural networks (NN), and SVM, etc. In those researches, all of DT ensemble studies have demonstrated impressive improvements in the generalization behavior of DT, while NN and SVM ensemble studies have not shown remarkable performance as shown in DT ensembles. Recently, several works have reported that the performance of ensemble can be degraded where multiple classifiers of an ensemble are highly correlated with, and thereby result in multicollinearity problem, which leads to performance degradation of the ensemble. They have also proposed the differentiated learning strategies to cope with performance degradation problem. Hansen and Salamon (1990) insisted that it is necessary and sufficient for the performance enhancement of an ensemble that the ensemble should contain diverse classifiers. Breiman (1996) explored that ensemble learning can increase the performance of unstable learning algorithms, but does not show remarkable performance improvement on stable learning algorithms. Unstable learning algorithms such as decision tree learners are sensitive to the change of the training data, and thus small changes in the training data can yield large changes in the generated classifiers. Therefore, ensemble with unstable learning algorithms can guarantee some diversity among the classifiers. To the contrary, stable learning algorithms such as NN and SVM generate similar classifiers in spite of small changes of the training data, and thus the correlation among the resulting classifiers is very high. This high correlation results in multicollinearity problem, which leads to performance degradation of the ensemble. Kim,s work (2009) showedthe performance comparison in bankruptcy prediction on Korea firms using tradition prediction algorithms such as NN, DT, and SVM. It reports that stable learning algorithms such as NN and SVM have higher predictability than the unstable DT. Meanwhile, with respect to their ensemble learning, DT ensemble shows the more improved performance than NN and SVM ensemble. Further analysis with variance inflation factor (VIF) analysis empirically proves that performance degradation of ensemble is due to multicollinearity problem. It also proposes that optimization of ensemble is needed to cope with such a problem. This paper proposes a hybrid system for coverage optimization of NN ensemble (CO-NN) in order to improve the performance of NN ensemble. Coverage optimization is a technique of choosing a sub-ensemble from an original ensemble to guarantee the diversity of classifiers in coverage optimization process. CO-NN uses GA which has been widely used for various optimization problems to deal with the coverage optimization problem. The GA chromosomes for the coverage optimization are encoded into binary strings, each bit of which indicates individual classifier. The fitness function is defined as maximization of error reduction and a constraint of variance inflation factor (VIF), which is one of the generally used methods to measure multicollinearity, is added to insure the diversity of classifiers by removing high correlation among the classifiers. We use Microsoft Excel and the GAs software package called Evolver. Experiments on company failure prediction have shown that CO-NN is effectively applied in the stable performance enhancement of NNensembles through the choice of classifiers by considering the correlations of the ensemble. The classifiers which have the potential multicollinearity problem are removed by the coverage optimization process of CO-NN and thereby CO-NN has shown higher performance than a single NN classifier and NN ensemble at 1% significance level, and DT ensemble at 5% significance level. However, there remain further research issues. First, decision optimization process to find optimal combination function should be considered in further research. Secondly, various learning strategies to deal with data noise should be introduced in more advanced further researches in the future.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.20 no.12
• /
• pp.844-861
• /
• 2008

The papers published in the Korean Journal of Air-Conditioning and Refrigeration Engineering during the year of 2007 have been reviewed. Focus has been put on current status of research in the aspect of heating, cooling, ventilation, sanitation and building environments. The conclusions are as follows. (1) The research trends of fluid engineering have been surveyed as groups of general fluid flow, fluid machinery and piping, etc. New research topics include micro nano fluid, micropump and fuel cell. Traditional CFD was still popular and widely used in research and development. Studies about fans and pumps were performed in the field of fluid machinery. Characteristics of flow and fin shape optimization are studied in the field of piping system. (2) The research works on heat transfer have been reviewed in the field of heat transfer characteristics, heat exchangers, and desiccant cooling systems. The research on heat transfer characteristics includes thermal transport in pulse tubes, high temperature superconductors, ground heat exchangers, fuel cell stacks and ice slurry systems. For the heat 'exchangers, the research on pin-tube heat exchanger, plate heat exchanger, condensers and gas coolers has been cordially implemented. The research works on heat transfer augmenting tubes have been also reported. For the desiccant cooling systems, the studies on the design and operating conditions for desiccant rotors as well as performance index are noticeable. (3) In the field of refrigeration, many papers were presented on the air conditioning system using CO2 as a refrigerant. The issues on the two-stage compression, the oil selection, and the appropriate oil charge were treated. The subjects of alternative refrigerants were also studied steadily. Hydrocarbons, DME and their mixtures were considered and various heat transfer correlations were proposed. (4) Research papers have been reviewed in the field of building facilities by grouping into the researches on heat and cold sources, air conditioning and air cleaning, ventilation and fire research including tunnel ventilation, flow control of piping system, and sound research with drain system. Main focuses have been addressed to the promotion of efficient or effective use of energy, which helps to save energy and results in reduced environmental pollution and operating cost. (5) Studies were mostly focused on analyzing the indoor environment in various spaces like cars, old tombs, machine rooms, and etc. in an architectural environmental field. Moreover, subjects of various fields such as the evaluation of noise, thermal environment, indoor air quality and development of energy analysis program were researched by various methods of survey, simulation, and field experiment.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.4 s.304
• /
• pp.1-12
• /
• 2005

Image registration is a process to establish the spatial correspondence between images of the same scene, which are acquired at different view points, at different times, or by different sensors. This paper presents a new algorithm for robust registration of the images acquired by multiple sensors having different modalities; the EO (electro-optic) and IR(infrared) ones in the paper. The two feature-based and intensity-based approaches are usually possible for image registration. In the former selection of accurate common features is crucial for high performance, but features in the EO image are often not the same as those in the R image. Hence, this approach is inadequate to register the E0/IR images. In the latter normalized mutual Information (nHr) has been widely used as a similarity measure due to its high accuracy and robustness, and NMI-based image registration methods assume that statistical correlation between two images should be global. Unfortunately, since we find out that EO and IR images don't often satisfy this assumption, registration accuracy is not high enough to apply to some applications. In this paper, we propose a two-stage NMI-based registration method based on the analysis of statistical correlation between E0/1R images. In the first stage, for robust registration, we propose two preprocessing schemes: extraction of statistically correlated regions (ESCR) and enhancement of statistical correlation by filtering (ESCF). For each image, ESCR automatically extracts the regions that are highly correlated to the corresponding regions in the other image. And ESCF adaptively filters out each image to enhance statistical correlation between them. In the second stage, two output images are registered by using NMI-based algorithm. The proposed method provides prospective results for various E0/1R sensor image pairs in terms of accuracy, robustness, and speed.

• The korean journal of orthodontics
• /
• v.41 no.5
• /
• pp.312-323
• /
• 2011

Objective: This study compared the bone thickness of the palate between lateral cephalogram and 3D model measurements. Methods: The subjects consisted of 30 adults (15 men,15 women) with a normal skeletal pattern and occlusion. The CT images were transformed to a 3D model, and were compared with the cephalometric image. Descriptive statistics for each variable were calculated. Results: In the 3D CT model, the mid-palatal area was the thickest part. It became thinner as the palate tapered laterally. In the male group, the thinnest portion was positioned 6 mm away from the mid-palate, while in the female group the thinnest portion was 8mm away from the mid-palate. Correlation analysis between the lateral cephalometric and 3D CT model revealed a significant correlation except in the mid palatal area and the area 2 mm lateral to the mid-palate in men, whereas there was a significant relationship in every area in the women. In both men and women, the highest correlation appeared in the area 8 mm lateral to the mid palate. Conclusions: Using regression analysis, an actual prediction of the bone thickness between the measured bone thickness of the lateral cephalometric radiograph and 3D model was made. This will provide useful information for mini-implant length selection when inserting into the palate.

• Proceedings of the KOR-BRONCHOESO Conference
• /
• 1981.05a
• /
• pp.37.1-37
• /
• 1981

Even through the methods of improving hearing are widely practiced in Korea, with its 30 years of history, there are many different opinions regarding the selection of surgical technique and the materials to be used in tympanoplasty. Also there are varying standards of postoperative evaluations. Therefore, we have done research to evaluate hearing improvement which in one of the objective of the middle ear surgery. The research was conducted for one year, January to December 1980. This research compared the level of hearing improvement after tympanoplasty, according to materials used in tympanoplasty and collumelization. Following are the data we have obtained; 1) Total number of cases we have reviewed were 306. Out of this, 35% of the cases were tympanoplasty type 1, 12.7% of the cases were collumelization with mastoidectomy, and 11.9% of the cases received collumelization without mastoidectomy. 2) We have conducted audiometry on 41.1% of the tympanoplasty type 1, 64.1% of the collumelization with mastoidectomy, and 45.7% of the collumelization without mastoidectomy. 3) We have observed above 11 dB hearing improvement in 70% of the tympanoplasty type 1, 36% of the collumelization with mastoidectomy, and 44% of the collumelization without mastoidectomy. 4) Over 11 dB decrease of air-bone gap in 61% of the tympanoplasty type 1,32% of the collumelization with mastoidectomy, and 63% of the collumelization without mastoidectomy. 5) If we look at the hearing improvement according to the materials used in the tympanoplasty, there was above 11 dB improvement of the air conduction in 63% with cartilage and 54% with fascia. In air-bone gap, 56% with cartilage and 52% with fascia. 6) If we look at the hearing improvement according to the materials used in the collumelization, there was above 11 dB improvement in 50% with cartilage, 14% with homograft, 55% with autograft. In air-bone gap, 56%, 21%, and 55% respectively.

• Journal of Korean Home Economics Education Association
• /
• v.24 no.2
• /
• pp.135-154
• /
• 2012

The purpose of this study was to investigate the trends of HE exam questions through analyzing the examination items for secondary Home Economics(HE) teaching certification from 2002 to 2011. The results of the study were as follows: First, regarding the analysis on pedagogy of HE course, it accounted for 30.9% of the total questions, and recently, it increased to 35%. Regarding the ratio of questions by each evaluation element, 'practice of teaching and learning methods for HE course'(33.2%) had the highest ratio, and 'essence of the education of HE'(3.2%) had the lowest ratio. Second, regarding the analysis on eating habits, it amounted to 17.5% of the total questions, and recently reached 15%. With regard to the ratio of each evaluation element, 'cooking theory and practice'(26.1%) had the highest ratio, and 'culture of eating habits'(3.7%) had the lowest ratio. Third, regarding the analysis on clothing habits, it accounted 15% of the total questions, and recently reached 12.5%. With regard to the ratio of each evaluation element, 'clothing management'(25.1%) and 'production of clothing and living necessaries'(25.1%) had the highest ratio, and 'selection of clothing and self-expression'(3.0%) had the lowest ratio. Fourth, regarding the analysis on housing life, it accounted for 11% of the total questions, and recently reaches 12.5%. With regard to each evaluation element, 'the understanding of housing life culture'(22.5%) had the highest ratio, and 'the understanding of housing interior design'(10.7%) had the lowest ratio. Fifth, regarding the analysis on daily life as a consumer, it accounted for 12.4%, and recently reaches 12.5%. With regard to the ratio of each evaluation element 'management of domestic resources'(34%) had the highest ratio, and 'planning of entire life and domestic welfare'(0%) had the zero ratio. Sixth, regarding the analysis on family life, it accounted for 13.3% of the total questions, and recently reaches 12.5%. With regard to the ratio of each evaluation element, 'the understanding of family and the changes in family'(23.8%) had the highest ratio, and 'marriage and the development of family'(2%) had the lowest ratio.

• Journal of the Korea Institute of Building Construction
• /
• v.9 no.4
• /
• pp.119-129
• /
• 2009

Building construction trends have been changed dramatically in terms of size and mass. With the need to maximize land usage, there has been an increase in the construction of high-rise buildings. This affects not only the entire construction duration and cost, but also subsequent construction activities, such as work to increase underground facilities and in reclamation land area construction. These types of site conditions require soft ground reinforcement and the proper uplift water pressure treatment. In general, two kinds of methods have been used for uplift water pressure treatment systems. However, there have been some problems arising as the result of a lack of research and analysis on underground construction techniques, and a reliance on experiments over actual survey and analysis of site conditions. This paper focused on the problems of conventional selection procedure, by analyzing drawings and proposing a kind of modeling for a reasonable procedure. The results were applied to OO project as a sample construction case to be verified in this research. The initial plan in the case project was the Rock Anchor System. However, as there were terrible miscalculations of basic site conditions that had an extraordinary influence on the underground water level, such as the site's proximity to the Han-river, it was necessary to change the plan to include apermanent drainage system. This achieved a direct construction cost reduction \ 406,702,000 and a maximum sayings of 4% of operational cost, based on the 50-year building Life Cycle Cost.

• The Korean Journal of Pesticide Science
• /
• v.15 no.1
• /
• pp.55-60
• /
• 2011

To work out quality control methods of environmental-friendly organic materials (EFOMs), the reason and basis for EFOM-selection and farmer's favorite formulation type of EFOMs, etc were investigated on farmers who had been practicing environmental-friendly agriculture. EFOMs used were soil amendments, control agents of plant diseases and insect pests, plant growth promotion formulations, in turns. In EFOMs application time, 22.7% of farmers sprayed EFOMs without delay after they were bought, in other hand, 77.3% of farmers used EFOMs which had been bought and stored for some period. Microbial density on seventeen environmental-friendly microbial formulates (EFMFs) including microbial pesticides, a microbial fertilizer, and environmental-friendly organic materials was investigated at different storing temperature and shelf life. When the microbial density of EFMFs was investigated without delay after they were bought, all used microbial pesticides and a microbial fertilizer was confirmed to be optimal for the certified density but two of environmental-friendly organic materials was confirmed not to be optimal. When microbial density of 17 EFMFs were investigated after storing them for six months at $4^{\circ}C$, only one of 9 microbial pesticides was confirmed not to be optimal, the other hand four of seven environmental-friendly organic materials not to be optimal, which each of their microbial density was less than the certified density. Population dynamics of microbial agents was much more influenced in fluctuated temperature (room temperature) than in static temperature condition ($5^{\circ}C$ and $25^{\circ}C$). Shelf life of microbial agents according to microbial formulation type were high in granule type, liquid wettable type and liquid type in turns.

• Journal of radiological science and technology
• /
• v.28 no.2
• /
• pp.105-110
• /
• 2005

Background/Aim : Common bile duct (CBD) stones may cause jaundice, cholangitis, or pancreatitis. Extracorporeal shock wave lithotripsy (ESWL) may be needed whenever endoscopic procedure are failed to extract common bile duct stones. The aim of this study is to provide the standard for patient's best choice on ESWL for treatment of CBD stones resistant to endoscopic extraction. Materials and Methods : Fourty-six patients failed in endoscopic stone extraction including mechanical lithotripsy were treated by ESWL. In all patients, endoscopic sphincterotomy and nasobiliary drainage tube was done before ESWL using the ultrasonography for stone localization with a spark-gap type lithotriptor. Patients were sedated with an intravenous injection of 50 mg of Demerol. None were treated under general anesthesia. Results : Overall complete clearance rate of CBD stone was 89.1% (41/46). In 82.6% of the patients, the stones were extracted endoscopically after ESWL, and spontaneous passage was observed in 6.5%. In the clearance rate after ESWL, there were no noticeable differences with regard to number (single: 82.8%, two or three: 100%, more than three: 100%) and size of the stone (less than 33mm: 92.9%, 33 mm or larger: 83.3%), whereas there were significant differences with regard to the ratio of sum of long-axis length of the all stones to sum of long-axis length of the CBD excluding stone (1:2.4, 1:2.1) and diameter of the largest stone to diameter of CBD excluding stone (1:0.9, 1:0.4) for patients with complete clearance compared with those without. Conclusion : We propose that stones without the fragments are travelable sufficient space in CBD or extractable sufficient diameter of CBD regardless of stone size and number should be treated by other technique to prevent time and cost consuming, such as percutaneous transhepatic cholangioscopylithotomy.

• Water for future
• /
• v.5 no.1
• /
• pp.27-36
• /
• 1972

This thesis is the study of the rainfall probability depth in the major areas of Korea, such as Seoul, Pusan and Taegu. The purpose of the paper is to analyze the rainfall in connection with the safe planning of the hydraulic structures and with the project life. The methodology used in this paper is the statistical treatment of the rainfall data in the above three areas. The scheme of the paper is the following. 1. The complementation of the rainfall data We tried to select the maximm values among the values gained by the three methods: Fourier Series Method, Trend Diagram Method and Mean Value Method. By the selection of the maximum values we tried to complement the rainfall data lacking in order to prevent calamities. 2. The statistical treatment of the data The data are ordered by the small numbers, transformed into log, $\sqrt{}, \sqrt[3]{}, \sqrt[4], and$\sqrt[5], and calculated their statistical values through the electronic computer. 3. The examination of the distribution types and the determination of the optimum distibution types By the $x^2-Test$ the distribution types of rainfall data are examined, and rejected some part of the data in order to seek the normal rainfall distribution types. In this way, the optimum distribution types are determined. 4. The computation of rainfall probability depth in the safety project life We tried to study the interrelation between the return period and the safety project life, and to present the rainfall probability depth of the safety project life. In conclusion we set up the optimum distribution types of the rainfall depths, formulated the optimum distributions, and presented the chart of the rainfall probability depth about the factor of safety and the project life.ct life.