• 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.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.18 no.3
• /
• pp.145-155
• /
• 2002

The results of the present feasibility study are summarized as follows, 1. The three unit bridge of knitted material and UD fibre reinforcement has both the rigidity and the strength against a vertical occlusal load of 75N. 2. Stress concentration at the junctional area between the bridge and the abutments, i.e. between the pontic and the knitted caps was observed. In the case of the bridge with reinforcement straps, it was partly shown that the concentration problem could be improved by simply increasing the fillet size at the area. Further refining in the surface of the junctional area will be needed to ensure a further improvement in the stress distribution. This will require some trade off in the level of the stress and the available space. A parametric study will help to decide the appropriate size of the fillet. 3. Design refinement is a must to improve the stress distribution and realize the most favourable shape in terms of fabrication. The current straight bar with a constant cross section area can be redesigned to a tapered shape. The curve from the dental arch should also be placed on the pontic design. In accordance with design refinement, the resistance of the bridge frame to other load cases should be evaluated. 4. Although not included in the present feasibility study, it is estimated that bridges of the anterior teeth can be made strong enough with the knitted material without further reinforcement using unidirectional materials. In this regard, a feasibility study on design concepts and stress analysis for 3, 4, 5 unit bridge is suggested. 5. Two types of bridge were analysed in terms of fatigue. The safe life design concept, i.e. fatigue design concept, looks reasonable for the bridge where if cracks should form and propagate there is virtually nothing a dentist to do. The bridge must be designed so that no crack will be initiated during the life span. In the case of crowns, however, if constructed with composite resin with knitted materials, it might be possible to repair them, which in general is impossible for crowns of PFM or of metal. Therefore for composite resin crowns, a damage tolerance design concept can be applied and reasonably higher operational stresses can be allowed. In this case, of course, a periodic inspection program should be established in parallel. 6. Parts of future works in terms of structural viewpoint which need to be addressed are summarized as the following: 1) To develop processing technology to accommodate design concepts; 2) More realistic modelling of the bridge and analysis-geometry and loading condition. Thickness variation in the knitted material, taper in the pontic, design for anterior tooth bridge, the effect of combined loads, etc, will need to be included; 3) To develop appropriate design concepts and design goals for the fibre composite FPD aiming at taking the best advantage of knitted materials, including the damage tolerance design concept; 4) To develop testing method and perform test such as static ultimate load test, fatigue test, repair test, etc, as necessary.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.37 no.2
• /
• pp.14-25
• /
• 2009

We live in an age which is exponentially growing as the knowledge paradigm is changing. New sites are subject to contemporary landscape architecture function as "fields" in which this hybrid aspect is both actively practiced and becoming a catalyst for change in the area of landscape architecture. With this as its background, this study attempts to deal with how the aspect of integration in environmental design is manifested. For this purpose, the tendencies for the discussion of integration in various fields of practice were examined: planning theories, urban theories, architecture, public environment, engineering, and landscape architecture. As yet, the discussions of interdisciplinary integration, which occur in practice in these respective fields, mainly tend to be oriented toward the effective implementation of the merits of other related fields. Seen from these examples of practice, integrated design approaches can be found in the following three aspects: design objects, respective professional areas, and methodologies of approaches and design. In terms of design objects, the positions of individual design subjects present themselves as most obvious, and integration or combination of the physical targets that come to exist through design can be easily seen. Most examples of integration turn out to be this, in almost every case of which the theme and the target of expression are integrated via a small number of certain methods. In terms of professional areas, what can be mainly evidenced is how the individual subject acts when the subject designs. The strong points of professionals from each field seem to create synergy, achieving through integration optimum results. In terms of methodologies of approaches and design, there are attempts to create integrated approaches as ways of effective decision-making, in which case the integration of all of the interest parties is of primary concern. As yet, few instances have been found in which integrated design has had enough strength to be seen as a concrete design methodology based on practical examples. However, it is encouraging that theoretical approaches and the necessity for integrated design have been identified from multiple perspectives, and that a practical movement such as landscape urbanism has come into active being. The authors of this study find this point in time to be ripe for discussions on integrated practices in terms of environmental design, on the basis of the synthetic approaches mentioned above.

• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.241-254
• /
• 2011

Financial time-series forecasting is one of the most important issues because it is essential for the risk management of financial institutions. Therefore, researchers have tried to forecast financial time-series using various data mining techniques such as regression, artificial neural networks, decision trees, k-nearest neighbor etc. Recently, support vector machines (SVMs) are popularly applied to this research area because they have advantages that they don't require huge training data and have low possibility of overfitting. However, a user must determine several design factors by heuristics in order to use SVM. For example, the selection of appropriate kernel function and its parameters and proper feature subset selection are major design factors of SVM. Other than these factors, the proper selection of instance subset may also improve the forecasting performance of SVM by eliminating irrelevant and distorting training instances. Nonetheless, there have been few studies that have applied instance selection to SVM, especially in the domain of stock market prediction. Instance selection tries to choose proper instance subsets from original training data. It may be considered as a method of knowledge refinement and it maintains the instance-base. This study proposes the novel instance selection algorithm for SVMs. The proposed technique in this study uses genetic algorithm (GA) to optimize instance selection process with parameter optimization simultaneously. We call the model as ISVM (SVM with Instance selection) in this study. Experiments on stock market data are implemented using ISVM. In this study, the GA searches for optimal or near-optimal values of kernel parameters and relevant instances for SVMs. This study needs two sets of parameters in chromosomes in GA setting : The codes for kernel parameters and for instance selection. For the controlling parameters of the GA search, the population size is set at 50 organisms and the value of the crossover rate is set at 0.7 while the mutation rate is 0.1. As the stopping condition, 50 generations are permitted. The application data used in this study consists of technical indicators and the direction of change in the daily Korea stock price index (KOSPI). The total number of samples is 2218 trading days. We separate the whole data into three subsets as training, test, hold-out data set. The number of data in each subset is 1056, 581, 581 respectively. This study compares ISVM to several comparative models including logistic regression (logit), backpropagation neural networks (ANN), nearest neighbor (1-NN), conventional SVM (SVM) and SVM with the optimized parameters (PSVM). In especial, PSVM uses optimized kernel parameters by the genetic algorithm. The experimental results show that ISVM outperforms 1-NN by 15.32%, ANN by 6.89%, Logit and SVM by 5.34%, and PSVM by 4.82% for the holdout data. For ISVM, only 556 data from 1056 original training data are used to produce the result. In addition, the two-sample test for proportions is used to examine whether ISVM significantly outperforms other comparative models. The results indicate that ISVM outperforms ANN and 1-NN at the 1% statistical significance level. In addition, ISVM performs better than Logit, SVM and PSVM at the 5% statistical significance level.

• Journal of the Korean Society of Radiology
• /
• v.7 no.6
• /
• pp.403-408
• /
• 2013

This study made a tentative estimation of the shielding rate of barium compound by thickness through monte carlo simulation to apply medical radiation shielding products that can replace existing lead. Barium sulfate($BaSO_4$) was used for the shielding material, and thickness of the shielding material specimen was simulated from 0.1 mm to 5 mm by applying $15{\times}15cm^2$ of specimen area, $4.5g/cm^3$ of density of barium sulfate, and $11.34g/cm^3$ density of lead. Entered source was simulated with 10kVp Step in consecutive X-ray energy spectrum(40 kVp ~ 120 kVp). Absorption probability in 40 kVp ~ 60 kVp showed same shielding rate with lead in 3 mm ~ 5 mm of thickness, but it was identified that under 2 mm, the shielding rate was a bit lower than the existing lead shielding material. Also, the shielding rate in 70 kVp ~ 120 kVp energy band showed similar performance as the existing lead shielding material, but it was tentatively estimated as fairly low shielding rate below 0.5 mm. This study estimated the shielding rate of barium compound as the thickness function of x-ray energy band for medical radiation through monte carlo simulation, and made comparative analysis with existing lead. Also, this study intended to verify application validity of the x-ray shielding material for medical radiation of pure barium sulfate. As a result, it was estimated that the shielding effect was 95% higher than the existing lead 1.5 mm in at least 2 mm thickness of barium compound in medical radiation energy band 70 kVp ~ 120 kVp, and this result is considered valid to be provided as a base data in weight lightening production of radiation shielding product for medical radiation.

• Korean Journal of Agricultural Science
• /
• v.9 no.1
• /
• pp.239-249
• /
• 1982

The outlines of superior stands, including species mixed, age classes, canopies, and ecological conditions of the stands, were examined in the forests of west Germany, Sweden, and Netherland. The establishment schemes of tree seed orchards for the production of high genetic quality seed were also investigated in the present study. 1. The meteorological conditions, abundant in precipitation and high in relative humidity, prove it to be oceanic climate characteristics in the Scandinavia located in high latitude and north-western part of Europe. The photo periodical characteristics following to day-length and the frost times determine the patterns of tree growth and distribution as well as the tree adaptability due to bud-setting and bud-flushing in Sweden forest. 2. The European forests are generally characterized by dense multistoried forest type composed of upper stories of tall tree species such as Norway spruce, fir, Scots pine and beech, lower stories of shade tolerant poles and saplings of spruce, fir, beech, and floor vegetations covered with birch, oak and hornbeam etc. 3. The single tree selection method and group shelterwood method are commonly applied for silvicultural system but shelterwood method and clearcutting method practiced partially in limited area. Badischer Femelschlagbetrieb and Bayerischer Femelschlagbetrieb are popularized to the regeneration of Schwarzwald but selection method applied in part. 4. Tree cuttings are properly limited so as not to exceed annual increment volume for the sustained yield and the preservation of ecosystem in European forest management. Land preservation and recreation function are more interested with the promotion of wood production in the forest management. 5. Careful attentions are paid to the reforestation of the introduced exotic species, such as Douglas-fir, red oak, pines from America, Japanese larch, Italian poplar etc., with concern against the insect and pest damage, and the meteorological damage. 6. Some intensive studies on tree improvement, such as selection of plus tree, selection of plus stand for seed collection, provenance trial, progeny test of clone, hybrid seed orchard for clone complexing, are performed in the countries and the great part of the seed required for reforestation are supplied from the seed orchard established in Sweden and Netherland.

• Horticultural Science & Technology
• /
• v.30 no.5
• /
• pp.543-548
• /
• 2012

Morphological characters of Baemoochae, xBrassicoraphanus are mostly intermedium of the both parents, Chinese cabbage, Brassica rapa ssp. pekinensis and radish, Raphanus sativus. The upper and lower parts of the leaf resemble the shape of Chinese cabbage and radish, respectively. The midrib of the leaf is round like to that of radish, but very big more than 3 cm in diameter and white in color like that of Chinese cabbage. The root was changed from the swollen type like that of radish to the enlarged taproot like that of the land race of Chinese cabbage after attaining genetical stability. The flower is white. The seed pod is divided into 2 different parts; the upper part is radish and about 4 cm in length and holds 3-4 seeds and the lower part is Chinese cabbage and about 3 cm in length and holds 7-8 seeds. The color of seed is brown, weight per 1.000 seeds is 5.5 g and the number of seeds per mL is 120. The matured plant in the fall season is around 5 kg in weight and outer leaves are very vigorous and stiffly and inner leaves are erect and form a loose head. The leaf and the root contain a high level of sulforaphene which is well known as a functional substance for anti-cancer and anti-super-bacteria. Baemoochae is an amphidiploid and does not have the self incompatibility function. It has a high level of cross compatibility with Chinese cabbage as the female parent, but not the male parent. It is cross incompatible to cabbage, B. oleracea, black mustard, B. nigra and radish. However it is highly compatible to oil seed rape, B. napus, yellow mustard, B. carinata and partial compatible to muatard, B. juncea in the reciprocal cross.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.35 no.2
• /
• pp.136-146
• /
• 1999

The telemetry system for the oxygen, pH, turbidity and the distribution ecology of fishes was constructed by the authors in order to product and manage effectively in shallow sea culture and setnets fisheries, and then the experiments for the telemetry system carried out at the culturing fishing ground in coast of Sanyang-Myon, Kyoungsangnam-Do and the set net fishing ground located Nungpo bay in Kojedo province respectively from October, 1997 to June 1998.As those results, the techniques suggested in the telemetry system for which find out the relationship between the physical and chemical environment in the sea and the distribution ecology of fishes gave full display its function, and its system could be operated as real time system. This research can also provide base-line data to develope a hybrid system unifying the marine environment information and the fisheries resources information in order to manage effectively coastal fishing ground.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.28 no.4
• /
• pp.371-383
• /
• 2012

Shaping the root canal system to maintain original canal curvature is essential to clinical success in endodontic treatment. Opposed to most root canals that are curved, endodontic instruments are made from straight metal blanks. They have a tendency of straightening the root canal during preparation and frequently result in procedural errors. A new treatment method to maintain original canal curvature during shaping has been introduced for preventing procedural errors. The aim of this study was to compare the ability of instruments to maintain original canal curvature of continuous rotary system and single file system. Thirty ISO 15, 0.02 taper, Endo Training Blocks(Dentsplay Maillefer) were used. Specimens were assigned to 1 of 3 groups for shaping: specimens in group 1 were shaped with ProFile #20/.06 at the WL. Specimens in group 2 were shaped with Mtwo #35/.04 at the WL. Specimens in group 3 were shaped with WaveOne Primary reciprocating files at the WL after the glide path was achieved with PathFile. Pre- and postinstrumentation digital images were superimposed and processed with Matlab r2010b(The MathWorks Inc, Natick, MA) software to analyze the curvature-radius ratio(CRr), representing canal curvature modification. Data for comparison on the ability of instruments to maintain original canal curvature depending on each Ni-Ti file were analyzed with 1-way ANOVA(P<.05). Data for comparison on the ability of instruments to maintain original canal curvature depending on each Ni-Ti file system were analyzed with independent t-test(P<.05). A statistically significant difference(P<0.05) was noted on each Ni-Ti file. ProFile and WaveOne instrumentations maintained the original canal curvature significantly better(P<0.05) than Mtwo file. There were no significant difference(P>0.05) between continuous rotary system and single file system. Under the conditions of this study, ProFile and WaveOne instruments maintained the original curvature significantly better than Mtwo file and were less modification of the canal curvature compared. There was no significant difference between continuous rotary system and single file system in shaping of simulated canals. As clinical practitioners, it may be advantages to use hybrid approach when root canal shapes depending on the design and usage of Ni-Ti files.

• Korean Journal of Community Nutrition
• /
• v.26 no.5
• /
• pp.327-336
• /
• 2021

Objectives: This study was conducted to create a 3D printable snack dish model for the elderly with low food or fluid intake along with barriers towards eating. Methods: The decision was made by the hybrid-brainstorming method for creating the 3D model. Experts were assigned based on their professional areas such as clinical nutrition, food hygiene and chemical safety for the creation process. After serial feedback processes, the grape shape was suggested as the final model. After various concept sketching and making clay models, 3D-printing technology was applied to produce a prototype. Results: 3D design modeling process was conducted by SolidWorks program. After considering Dietary reference intakes for Koreans (KDRIs) and other survey data, appropriate supplementary water serving volume was decided as 285 mL which meets 30% of Adequate intake. To consider printing output conditions, this model has six grapes in one bunch with a safety lid. The FDM printer and PLA filaments were used for food hygiene and safety. To stimulate cognitive functions and interests of eating, numbers one to six was engraved on the lid of the final 3D model. Conclusions: The newly-developed 3D model was designed to increase intakes of nutrients and water in the elderly with dementia during snack time. Since dementia patients often forget to eat, engraving numbers on the grapes was conducted to stimulate cognitive function related to the swallowing and chewing process. We suggest that investigations on the types of foods or fluids are needed in the developed 3D model snack dish for future studies.