• Journal of the Mineralogical Society of Korea
• /
• v.30 no.4
• /
• pp.161-172
• /
• 2017

Clay minerals play a major role in the geochemical cycles of metals in the Critical Zone, the Earth surface-layer ranging from the groundwater bottom to the tree tops. Atomistic scale research of the very fine particles can help understand the fundamental mechanisms of the important geochemical processes and possibly apply to development of hybrid nanomaterials. Molecular dynamics (MD) simulations can provide atomistic level insights into the crystal structures of clay minerals and the chemical reactivity. Classical MD simulations use a force field which is a parameter set of interatomic pair potentials. The ClayFF force field has been widely used in the MD simulations of dioctahedral clay minerals as the force field was developed mainly based on dioctahedral phyllosilicates. The ClayFF is often used also for trioctahedral mineral simulations, but disagreement exits in selection of the interatomic potential parameters, particularly for Mg atom-types of the octahedral sheet. In this study, MD simulations were performed for trioctahedral clay minerals such as brucite, lizardite, and talc, to test how the two different Mg atom types (i.e., 'mgo' or 'mgh') affect the simulation results. The structural parameters such as lattice parameters and interatomic distances were relatively insensitive to the choice of the parameter, but the vibrational power spectra of hydroxyls were more sensitive to the choice of the parameter particularly for lizardite.

• Journal of Korean Society of Forest Science
• /
• v.31 no.1
• /
• pp.8-20
• /
• 1976

For study on the variation of natural stand, three populations of Pinus densiflora S. et Z. were selected at samsanri Yongogmyun Myongjugun Kangwondo (4), Hawonri seomyun Uljingun Kyongbuk (5), and Emogdong Suweon Kyongkido (6) successibely after the selection of three population in 1974. Twenty individual trees were chosen from each population and the morphological characteristics of trees, needle and wood properties were investigated on the trees. The results are summerized as follows; 1. Serration density, resin canal number in needle did not show significant differences, however stomata row number in the both sides of needle showed highly significant differences among 3 populations. But significant differences were calculated among individual trees in a population regarding any character of needles. 2. Ail population had high correlation on the stomata row between abaxial and adaxial side of needle. 3. The Myongjungun population showed the highest value of resin duct index, which means the population had the highest degree of hybrid character. 4. The ring segment width and summerwood percentage in the wood properties had significant differences, and yet specific gravity and tracheid length had not significant differences statistically among 3 populations. But all the values were significant statistically among the ring segments within population. 5. The ring segment width decreased rapidly with increasing tree age but summerwood percentage, specific gravity, tracheid length increased slowly to the middle age of tree and then decreased slowly after the age. But the patterns of decrease or increase were some different by population. 6. The values of Uljingun population were generally high in the coefficient of variation on all the needle characters. And the values of Suweon population were always the highest and those of Myongjugun population the lowest in the coefficient of variation on all the wood properties.

• 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 Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.111-124
• /
• 2018

This study aims to develop a classification model for predicting the occurrence of hyperlipidemia, one of the chronic diseases. Prior studies applying data mining techniques for predicting disease can be classified into a model design study for predicting cardiovascular disease and a study comparing disease prediction research results. In the case of foreign literatures, studies predicting cardiovascular disease were predominant in predicting disease using data mining techniques. Although domestic studies were not much different from those of foreign countries, studies focusing on hypertension and diabetes were mainly conducted. Since hypertension and diabetes as well as chronic diseases, hyperlipidemia, are also of high importance, this study selected hyperlipidemia as the disease to be analyzed. We also developed a model for predicting hyperlipidemia using SVM and meta learning algorithms, which are already known to have excellent predictive power. In order to achieve the purpose of this study, we used data set from Korea Health Panel 2012. The Korean Health Panel produces basic data on the level of health expenditure, health level and health behavior, and has conducted an annual survey since 2008. In this study, 1,088 patients with hyperlipidemia were randomly selected from the hospitalized, outpatient, emergency, and chronic disease data of the Korean Health Panel in 2012, and 1,088 nonpatients were also randomly extracted. A total of 2,176 people were selected for the study. Three methods were used to select input variables for predicting hyperlipidemia. First, stepwise method was performed using logistic regression. Among the 17 variables, the categorical variables(except for length of smoking) are expressed as dummy variables, which are assumed to be separate variables on the basis of the reference group, and these variables were analyzed. Six variables (age, BMI, education level, marital status, smoking status, gender) excluding income level and smoking period were selected based on significance level 0.1. Second, C4.5 as a decision tree algorithm is used. The significant input variables were age, smoking status, and education level. Finally, C4.5 as a decision tree algorithm is used. In SVM, the input variables selected by genetic algorithms consisted of 6 variables such as age, marital status, education level, economic activity, smoking period, and physical activity status, and the input variables selected by genetic algorithms in artificial neural network consist of 3 variables such as age, marital status, and education level. Based on the selected parameters, we compared SVM, meta learning algorithm and other prediction models for hyperlipidemia patients, and compared the classification performances using TP rate and precision. The main results of the analysis are as follows. First, the accuracy of the SVM was 88.4% and the accuracy of the artificial neural network was 86.7%. Second, the accuracy of classification models using the selected input variables through stepwise method was slightly higher than that of classification models using the whole variables. Third, the precision of artificial neural network was higher than that of SVM when only three variables as input variables were selected by decision trees. As a result of classification models based on the input variables selected through the genetic algorithm, classification accuracy of SVM was 88.5% and that of artificial neural network was 87.9%. Finally, this study indicated that stacking as the meta learning algorithm proposed in this study, has the best performance when it uses the predicted outputs of SVM and MLP as input variables of SVM, which is a meta classifier. The purpose of this study was to predict hyperlipidemia, one of the representative chronic diseases. To do this, we used SVM and meta-learning algorithms, which is known to have high accuracy. As a result, the accuracy of classification of hyperlipidemia in the stacking as a meta learner was higher than other meta-learning algorithms. However, the predictive performance of the meta-learning algorithm proposed in this study is the same as that of SVM with the best performance (88.6%) among the single models. The limitations of this study are as follows. First, various variable selection methods were tried, but most variables used in the study were categorical dummy variables. In the case with a large number of categorical variables, the results may be different if continuous variables are used because the model can be better suited to categorical variables such as decision trees than general models such as neural networks. Despite these limitations, this study has significance in predicting hyperlipidemia with hybrid models such as met learning algorithms which have not been studied previously. It can be said that the result of improving the model accuracy by applying various variable selection techniques is meaningful. In addition, it is expected that our proposed model will be effective for the prevention and management of hyperlipidemia.

• Korean Journal of Breeding Science
• /
• v.41 no.2
• /
• pp.106-114
• /
• 2009

In maize, knowledge of genetic diversity and genetic relationships among elite inbred lines is an significant impact on the selection of parental lines for hybrid varieties. Genetic diversity and genetic relationships among 11 parental inbred lines of Korean waxy and normal corn varieties were analyzed using 50 SSR markers distributed over the whole genome. A total of 171 allele bands were detected with an average of 3.4 alleles per locus. Number of allele bands per locus ranged from two to six and gene diversity varied from 0.165 to 0.900 with an average of 0.596 depending on the SSR loci. The cluster tree recognized three major groups with 61.6% genetic similarity. Group I includes 7 inbred lines (KL103, HW1, HW4, HW6, HW7, HW8, HW9), with similarity coefficients of between 0.616 and 0.730. Group II includes 2 inbred lines (HF1, HF2), with similarity coefficients of 0.959. Group III includes 2 inbred lines (HW3, HW5), with similarity coefficients of 0.713. The present study indicates that the SSR markers chosen for this analysis are effective for the assessment of genetic diversity and genetic relationships among 11 parental inbred lines.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.23-46
• /
• 2021

Collaborative filtering, which is often used in personalization recommendations, is recognized as a very useful technique to find similar customers and recommend products to them based on their purchase history. However, the traditional collaborative filtering technique has raised the question of having difficulty calculating the similarity for new customers or products due to the method of calculating similaritiesbased on direct connections and common features among customers. For this reason, a hybrid technique was designed to use content-based filtering techniques together. On the one hand, efforts have been made to solve these problems by applying the structural characteristics of social networks. This applies a method of indirectly calculating similarities through their similar customers placed between them. This means creating a customer's network based on purchasing data and calculating the similarity between the two based on the features of the network that indirectly connects the two customers within this network. Such similarity can be used as a measure to predict whether the target customer accepts recommendations. The centrality metrics of networks can be utilized for the calculation of these similarities. Different centrality metrics have important implications in that they may have different effects on recommended performance. In this study, furthermore, the effect of these centrality metrics on the performance of recommendation may vary depending on recommender algorithms. In addition, recommendation techniques using network analysis can be expected to contribute to increasing recommendation performance even if they apply not only to new customers or products but also to entire customers or products. By considering a customer's purchase of an item as a link generated between the customer and the item on the network, the prediction of user acceptance of recommendation is solved as a prediction of whether a new link will be created between them. As the classification models fit the purpose of solving the binary problem of whether the link is engaged or not, decision tree, k-nearest neighbors (KNN), logistic regression, artificial neural network, and support vector machine (SVM) are selected in the research. The data for performance evaluation used order data collected from an online shopping mall over four years and two months. Among them, the previous three years and eight months constitute social networks composed of and the experiment was conducted by organizing the data collected into the social network. The next four months' records were used to train and evaluate recommender models. Experiments with the centrality metrics applied to each model show that the recommendation acceptance rates of the centrality metrics are different for each algorithm at a meaningful level. In this work, we analyzed only four commonly used centrality metrics: degree centrality, betweenness centrality, closeness centrality, and eigenvector centrality. Eigenvector centrality records the lowest performance in all models except support vector machines. Closeness centrality and betweenness centrality show similar performance across all models. Degree centrality ranking moderate across overall models while betweenness centrality always ranking higher than degree centrality. Finally, closeness centrality is characterized by distinct differences in performance according to the model. It ranks first in logistic regression, artificial neural network, and decision tree withnumerically high performance. However, it only records very low rankings in support vector machine and K-neighborhood with low-performance levels. As the experiment results reveal, in a classification model, network centrality metrics over a subnetwork that connects the two nodes can effectively predict the connectivity between two nodes in a social network. Furthermore, each metric has a different performance depending on the classification model type. This result implies that choosing appropriate metrics for each algorithm can lead to achieving higher recommendation performance. In general, betweenness centrality can guarantee a high level of performance in any model. It would be possible to consider the introduction of proximity centrality to obtain higher performance for certain models.

• Horticultural Science & Technology
• /
• v.34 no.4
• /
• pp.655-661
• /
• 2016

To develop scab-resistant pear (Pyrus spp.) varieties with fruits that are as crisp and juicy as Asian pears, a cross was made between 'Whangkeumbae' and 'Bartlett' varieties (P. pyrifolia ${\times}$ P. communis) at the Pear Research Institute of the National Institute of Horticultural & Herbal Science, Rural Development Administration, in 1994. Among the 285 seedlings, 'Greensis' was first selected in 2006 for its good eating quality and named in 2012 after regional adaptation tests in nine regions and ten experimental plots from 2007 to 2012. The tree showed a vigorous growth habit and semi-spreading characteristics, like 'Whangkeumbae'. The optimum fruit harvest date was also around Sept. 26 and fruit was round in shape and green in skin color at maturity. Average fruit weight was 470g, and the soluble solids content was $12.4^{\circ}Brix$. The flesh was very crisp and juicy, and had good eating quality. Its' leaf size was similar with 'Bartlett' and smaller than 'Whangkeumbae'. The average of full bloom date of 'Greensis' was determined as Apr. 26, which was six days later than 'Whangkeumbae' and similar with 'Bartlett'. S genotypes of 'Greensis' were identified as $S_4S_e$ by S-allele PCR product sequencing analysis. It seems that the $S_4$ allele was inherited from 'Whangkeumbae' and the Se allele from 'Bartlett'. 'Greensis' displayed strong resistance to scab disease caused by Venturia nashicola, similar to European pear cultivars like 'Beurre Hardy' and, 'Conference'. 'Greensis' was also highly resistant to black leaf spot (Alternaria kikuchiana) in the field

• Korean Journal of Food Science and Technology
• /
• v.49 no.6
• /
• pp.649-658
• /
• 2017

The potential of the ethyl acetate fraction from Pteridium aquilinum (EFPA) to improve the cognitive function in high-fat diet (HFD)-induced diabetic mice was investigated. EFPA-treatment resulted in a significant improvement in the spatial, learning, and memory abilities compared to the HFD group in behavioral tests, including the Y-maze, passive avoidance, and Morris water maze. The diabetic symptoms of the EFPA-treated groups, such as fasting glucose and glucose tolerance, were alleviated. The administration of EFPA reduced the acetylcholinesterase (AChE) activity and malondialdehyde (MDA) content in mice brains, but increased the acetylcholine (ACh) and superoxide dismutase (SOD) levels. Finally, kaempferol-3-o-glucoside, a major physiological component of EFPA, was identified by using high-performance liquid chromatography coupled with a hybrid triple quadrupole-linear ion trap mass spectrometer (QTRAP LC-MS/MS).

• The Plant Pathology Journal
• /
• v.19 no.5
• /
• pp.248-252
• /
• 2003

This study was carried out to evaluate potential of Shiranuhi cultivar against inoculation of causal pathogen of citrus canker, Xanthomonas axonopodis pv. citri in Jeju Island by comparing degree of susceptibility of fruits and leaves/twigs, and analysis of incidence of canker disease. Progression of symptom, disease incidence, and percent area of lesion were surveyed for evaluation of resistance. In greenhouse condition, symptoms of bacterial citrus canker progressed more rapidly in sweet orange, a moderately susceptible cultivar, than in other four cultivars (satsuma mandarin, 'Kiyomi', 'Shiranuhi' and 'Yuzu'). At 20 days after inoculation, disease severity was the highest in sweet orange (5.0$\pm$0.0), and all tested leaves were distorted or had dropped. On the other hand, 'Yuzu' showed the lowest disease severity (2.6$\pm$0.47), followed by 'Kiyomi' (4.0$\pm$0.0), 'Shiranuhi' (4.0$\pm$0.82), and satsuma mandarin (4.3$\pm$0.47). Percent area of lesion per leaf 30 days after inoculation was the highest in sweet orange (8.31$\pm$1.78), followed by satsuma mandarin (1.51$\pm$1.25), 'Shiranuhi' (1.39$\pm$0.94), and 'Kiyomi' (1.1$\pm$0.9), while the lowest was in 'Yuzu' (0.26$\pm$0.17). Infield condition, percentage of diseased leaf in 'Shiranuhi' was very low, 5.2$\pm$2.9, compared with sweet orange, 71.0$\pm$ 11.5, while that of satsuma mandarin and 'Kiyomi' were 6.9$\pm$7.0 and 4.3$\pm$2.0, respectively. Percentages of diseased leaf was higher (17.4$\pm$7.1) than that of diseased fruit (3.2$\pm$2.5) in severely diseased trees of Shiranuhi cultivar, and the disease was not observed on twig in open field condition. Lesion sizes on leaves and fruits in open field condition were 4.1$\pm$2.2 mm2 and 5.1$\pm$5.6 mm2, respectively, while those in greenhouse condition were 8.7$\pm$5.7 mm2, 10.4$\pm$9.2 mm2 and 5.6$\pm$2.6 mm2 on leaves, fruits and twigs, respectively. The disease was observed in 5.6% out of total 107 farmers Shiranuhi fields under polyethylene film house, and average percentages of diseased tree in 31 fields of Shiranuhi cultivar and adjacent satsuma mandarin fields were 0.02% and 14.8%, respectively. Average percentage of diseased fruit was 1.6% in satsuma mandarin which was not observed in anyone of all the 31 Shiranuhi farmers fields. Therefore, it was concluded that 'Shiranuhi' cultivar is not potential against causal pathogen of citrus canker disease in Jeju Island because the cultivar has similar resistance as satsuma mandarin which occupies over 95% of total 25,000 ha in Jeju Island in polyethylene film houses protected from outside.

• Korean Journal of Plant Resources
• /
• v.31 no.4
• /
• pp.396-418
• /
• 2018

This study was carried out to investigate the flora and the vegetation of Samusan mountain on Jecheon-si, located in the middle part of the Korean peninsula. The vascular plants which were collected in 9 times from June 2014 to October 2015 were identified as 502 taxa in total, including 102 families, 314 genera, 445 species, 6 subspecies, 49 varieties, 1 form and 1 hybrid. The largest families were as follows; Asteraceae (70 taxa), Poaceae (52 taxa), Rosaceae (30 taxa), Fabaceae (28 taxa), and Cyperaceae (20 taxa). Of them, Korean endemic plants numbered 10 taxa, and vascular plants listed in the red data according to the International Union for Conservation of Nature (IUCN) numbered 14 taxa. The floristic regional indicator plants found in this area were 61 taxa in total. Among them, 5 taxa revealed the floristic grade V, 11 for floristic grade IV, 14 for floristic grade III, 10 for floristic grade II, and 21 for floristic grade I. The alien plants were identified as 43 taxa and the percentage of naturalized index (NI) was 8.6%, and urbanization index (UI) was 13.4%, respectively. Samples of the forest vegetation on the Samusan Mt. were mainly classified as Pinus densiflora, Quercus variabilis, Q. acutissima, Q. mongolica, Zelkova serrata and Robinia pseudoacacia forest. The vertical structures of the forest were stable and the DBH-Class analyses showed that the dominant tree species would be maintained. In the surveyed areas, high plant diversity was shown, and a number of endemic, rare, calcicole plants and phytogeographically important plants were found. Nonetheless, numerous and diverse biological resources native have been consistently disturbed or damaged by human activities without some form of protection. Therefore, it is needed to set up strategies for conservation forest vegetation in this study area.