• The Korean Journal of Applied Statistics
• /
• v.37 no.5
• /
• pp.691-702
• /
• 2024

In this paper, we propose mixed projection forest (MPF), a new classification ensemble method that can be effectively applied in the field of big data analysis. When training individual classifiers within an ensemble, MPF uses oblique hyperplanes using combined rotation matrix derived from data projection techniques of principal component analysis (PCA) and canonical linear discriminant analysis (CLDA), thereby improving the accuracy of each classifier. Additionally, the diversity of individual classifiers is improved by generating various rotation matrices through random partitioning of the input variable set. This approach ultimately enhances classification performance and proves to be highly effective in big data analysis that demands precision. We conducted a performance comparison of MPF with existing classification ensemble models using 30 real or simulated datasets. The results indicate that MPF achieves competitive performance in terms of classification accuracy and classifier diversity.

• Journal of Information Processing Systems
• /
• v.11 no.2
• /
• pp.165-172
• /
• 2015

The inactive student rate is becoming a major problem in most open universities worldwide. In Indonesia, roughly 36% of students were found to be inactive, in 2005. Data mining had been successfully employed to solve problems in many domains, such as for educational purposes. We are proposing a method for preventing inactive students by mining knowledge from student record systems with several state of the art ensemble methods, such as Bagging, AdaBoost, Random Subspace, Random Forest, and Rotation Forest. The most influential attributes, as well as demographic attributes (marital status and employment), were successfully obtained which were affecting student of being inactive. The complexity and accuracy of classification techniques were also compared and the experimental results show that Rotation Forest, with decision tree as the base-classifier, denotes the best performance compared to other classifiers.

• Journal of Korean Society of Forest Science
• /
• v.107 no.1
• /
• pp.16-24
• /
• 2018

This study was investigated to select superior population and clones of Salix gracilistyla for promoting woody biomass resources through creating of short rotation coppices (SRC). Plant materials were collected from seven different population groups of S. gracilistyla and planted at two different nursery sites in Chuncheon and Yongin. Height and root collar diameter showed statistically significant differeces among population, nursery, and annual growth for three years (p<0.01). Aboveground biomass was harvested to investigate dry weight after 3 years from planting, the Youngwol population showed the highest yield as $4.8kg\;DW\;plant^{-1}$ in Chuncheon nursery which was more than double yield compared to the other nursery planted plants. However, Hongcheon and Wonju populations as $3.3kg\;plant^{-1}$, showed the best yield in Yongin nursery. In addition, there was a significant difference between collected populations and nurseries. But there was statistically significant different interaction between population (Pop) and nursery (Nur) (F value = 3.51, p<0.01). Therefore, the superior populations selected by this experiment could be cultivated as an excellent variety for woody biomass resources through the clonal test.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.9 no.2
• /
• pp.100-108
• /
• 2007

The Gwangneung KoFlux supersite, located in a rugged mountain region, is characterized by a low wind speed due to a mountain-valley circulation and rolling terrain. Therefore, it is essential to understand the effect of coordinate rotation on flux measurements by the eddy-covariance method. In this paper, we review the properties of three orthogonal coordinate frames (i.e., double, triple, and planar fit rotations) and apply to flux data observed at the Gwangneung supersite. The mean offset of vertical wind speed of sonic anemometer was inferred from the planar fit (PF) coordinate rotation, yielding the diurnal variation of about $\pm0.05ms^{-1}$. Double rotation $(\bar{v}=\bar{w}=0)$ produced virtually the same turbulent fluxes of heat, water, and $CO_2$ as those from the PF rotation under windy conditions. The former, however, resulted in large biases under calm conditions. The friction velocity, an important scaling parameter in the atmospheric surface layer, was more sensitive to the choice of coordinate rotation method.

• Journal of the Korean Society of Safety
• /
• v.20 no.2 s.70
• /
• pp.11-17
• /
• 2005

Rotational motion in the atmosphere around a fire may have a profound influence on the fire plume. This process underlies the occurrence of fire whirls. Fire whirls are rare but highly destructive phenomenon which were observed in a large forest, urban and building fires. The present study aims to investigate of the effect of rotation Strength on the fire whirl characteristics expeimentally. Experiments are performed for various sizes of fire source with different rotation strength. From the experimental observations, it is noted that the mean centerline temperature is gradually increased and mean radial temperature is decreased as increases rotation strength. The characteristic mean flame height of fire based on the visible observation is increased as increases of dimensionless swirl parameter, $\Omega/\alpha$, represented by swirl induced motion to buoyancy driven motion.

• Journal of Climate Change Research
• /
• v.7 no.4
• /
• pp.485-492
• /
• 2016

For reporting national greenhouse gas inventory in forest sector, the forest growing stock from the National Forest Inventory (NFI) system has used as activity data sources. The National Forest Inventory system was changed from rotation system by province to annual system by 5 years across the country. The forest growing stocks based on the new inventory system produced a different trend compared to the previous estimations. This study was implemented to recalculate previous forest growing stocks for time series consistency at a national level. The recalculation of forest growing stock was conducted in an overlap approach by the IPCC guideline. In order to support the more consistency data, we used calibration factors between applied stand volumes in 1985 and 2012, respectively. As a result, the time series of recalculated forest growing stock was to be consistency using the overlap approach and the calibration factor with the lower middle/middle site index. According to the applied overlap period, however, we will recalculate activity data using more complete data from national forest inventory system.

• Journal of Korean Society of Forest Science
• /
• v.97 no.1
• /
• pp.77-82
• /
• 2008

For the last few decades, forest resources in Korea have continuously increased in terms of their stocks. However, due to the low profitability of timber harvesting in Korea, their economic values have not been well recognized. Furthermore, their economic values have not been counted in national statistics associated with forest resources such as the Forestry Household Economy Survey or the National Accounting. This research fundamentally aimed at building a forest valuation system or a procedure that enables one to evaluate the national-level economic values of timber assets in Korea. For this research, it is necessarily required to understand the principles of appraisal methods and any issues raised in their practical applications. Thus, a comprehensive review of appraisal methods utilized in Korea and other countries was conducted in this research. Also, the current valuation system of the National Forests in Korea was investigated. Through the review and the investigation, it was found that an alternative consists of two appraisal methods, the 'Capitalized Income Value' method adopting the Internal Rate of Return (IRR) for the below-rotation age forest and the 'Derived Residual Value' method for the above-rotation age forest, could be the most acceptable for evaluating timber assets of forestry households derived from the Forestry Household Economy Survey in Korea.

• Journal of Korean Society of Forest Science
• /
• v.111 no.4
• /
• pp.613-621
• /
• 2022

The study aim was to investigate changes in the diameter, number of standing trees, stand volume per ha and site index by the forest survey order, climate zone (northern temperate, central temperate, southern temperate, and warm temperate regions), and altitude in 100 m intervals) by collecting samples of Robinia pseudoacacia from the fifth, sixth, and seventh national forest survey datasets. The rotation cutting age, which is a standard used for wood, was calculated. The changes were statistically analyzed by performing ANOVA and the Duncan multiple test. Diameter growth naturally increased according to the forest survey order and was lowest in the southern temperate region by climate zone and lowest at the 301-400 m altitude. The number of standing trees per ha did not change according to the forest survey order and altitude, and the density was highest in the central temperate region and lowest in the southern temperate region. The stand volume per ha increased according to the forest survey order, and the climate zone was divided into two groups: ① northern temperate region and central temperate region, ② southern temperate region and warm temperate region. The stand volume growth was highest at the 201-300 m point. Thesite index showed results similar to the change pattern of the stand volume per ha. The growth curve, which can be seen by the change in stand volume per ha, was estimated by applying theWeibull formula, and the stand volume per ha was estimated to reach approximately 200 m³/ha at 50-60 years. The rotation of the highest production in volume, which is the standard for using trees as wood rather than honey sources, was calculated to be 34 years.

• Journal of Climate Change Research
• /
• v.7 no.2
• /
• pp.137-142
• /
• 2016

To achieve large-scale carbon removals, a carbon offset project based on forest management was designed and its carbon stock change was estimated in this study. The project was designed for 874 hectares of forests in the Jin-An Leading Forest Management Zone. For estimating the carbon stock change of the project, the Korean Forest Carbon Standard and VCS (Verified Carbon Standard) methodologies were applied. Three types of management options were considered in the project : extension of rotation age, conversion to productive forests, and conversion to selective harvesting. The estimated carbon removals from the project designed in this study were $259,936tCO_2$ ($8,664tCO_2$ annually), which is 98% of estimated carbon removals from the entire 69 projects currently registered to the Forest Carbon Offset Registry in Korea. The results of this study showed that a large-scale carbon offset project based on forest management could have a huge potential to produce carbon offset credits.

• Journal of the Korean Society of Visualization
• /
• v.19 no.3
• /
• pp.63-68
• /
• 2021

In this study, data was augmentation through the SinGAN algorithm using small image data, and defects in rubber O-rings were detected using the random forest algorithm. Unlike the commonly used data augmentation image rotation method to solve the data imbalance problem, the data imbalance problem was solved by using the SinGAN algorithm. A study was conducted to distinguish between normal products and defective products of rubber o-ring by using the random forest algorithm. A total of 20,000 image date were divided into transit and testing datasets, and an accuracy result was obtained to distinguish 97.43% defects as a result of the test.