• Communications for Statistical Applications and Methods
• /
• v.16 no.5
• /
• pp.881-889
• /
• 2009

Support vector machines(SVMs) are known as flexible and efficient classifier of multivariate observations, producing a hyperplane or hyperdimensional curved surface in multidimensional feature space that best separates training samples by known groups. As various methodological extensions are made for SVM classifiers in recent years, it becomes more difficult to understand the constructed model intuitively. The aim of this paper is to visualize various SVM classifications tuned by several parameters in reduced dimensions, so that data analysts secure the tangible image of the products that the machine made.

• Journal of The Korean Astronomical Society
• /
• v.41 no.6
• /
• pp.157-161
• /
• 2008

We have estimated the fractal dimension of the molecular clouds associated with the Hii region Sh 156 in the Outer Galaxy. We selected the $^{12}CO$ cube data from the FCRAO CO Survey of the Outer Galaxy. Using a developed code within IRAF, we identified slice-clouds (2-dimensional clouds in velocity-channel maps) with two threshold temperatures to estimate the fractal dimension. With the threshold temperatures of 1.8 K, and 3 K, we identified 317 slice-clouds and 217 slice-clouds, respectively. There seems to be a turn-over location in fractional dimension slope around NP (area; number of pixel) = 40. The fractal dimensions was estimated to be D = $1.5\;{\sim}\;1.53$ for $NP\;{\geq}\;40$, where $P\;{\propto}\;A^{D/2}$ (P is perimeter and A is area), which is slightly larger than other results. The sampling rate (spatial resolution) of observed data must be an important parameter when estimating fractal dimension. Fractal dimension is apparently invariant when varying the threshold temperatures applied to slice-clouds identification.

• Proceedings of the KSRS Conference
• /
• 2002.10a
• /
• pp.411-416
• /
• 2002

The emergence of viral diseases transmitted is nowadays a central problem in the world. Problem, which is becoming very critical in developing countries, where the health systems are not yet enough developed to face the bursting of such diseases. Emerging viral diseases constitute one of the major threats to human being that are arising in the modern world. Besides bio-chemical and medical researches, new orientations are developed to understand the environmental dimensions of such emergence. Questions concerning the inter-plays between the environmental and disease dynamics are building up new investigations, both in remote sensing and GIS, for the elaboration of levels of organization of space and environment in relation to incidences, to gain understandings in these issues. Environmental attributes attached to land cover types: area, spatial heterogeneity and physical state, are derived from remote sensing and applied to uncover related dimensions of the Dengue disease.

• Kyungpook Mathematical Journal
• /
• v.47 no.3
• /
• pp.311-322
• /
• 2007

This paper deals with space Beltrami system with three characteristic matrices in even dimensions, which can be regarded as a generalization of space Beltrami system with one and two characteristic matrices. It is transformed into a nonhomogeneous $p$-harmonic equation $d^*A(x,df^I)=d^*B(x,Df)$ by using the technique of out differential forms and exterior algebra of matrices. In the process, we only use the uniformly elliptic condition with respect to the characteristic matrices. The Lipschitz type condition, the monotonicity condition and the homogeneous condition of the operator A and the controlled growth condition of the operator B are derived.

• International Journal of Aeronautical and Space Sciences
• /
• v.14 no.4
• /
• pp.334-340
• /
• 2013

This paper deals with the optimization of blade stiffened composite panels. The main objective of the research is to make response surfaces for the constraints. The response surface for warping thermal deformation was previously made for a fixed dimension composite structure. In this study, the dimensions of the blade stiffener were treated as design variables. This meant that a new response surface technique was required for the constraints. For the response surfaces, the lamination parameters, linear thermal expansions and dimensions of the structures were used as variables. A genetic algorithm was adopted as an optimizer, and an optimal result, which satisfied two constraints, was obtained. As a result, a new response surface was obtained, for predicting warping thermal deformation.

• International Journal of CAD/CAM
• /
• v.9 no.1
• /
• pp.25-29
• /
• 2010

Let $S_n^r(\Omega)$ be the spline space of degree n and smoothness r with respect to $\Omega$ where $\Omega$ is a triangulation of a planner polygonal domain. Dimensions of $S_n^r(\Omega)$ over the so-called unconstricted triangulation were given by Farin in [J. Comput. Appl. Math. 192(2006), 320-327]. In this paper, a counter example is given to show that the condition used in the main result in Farins paper is not correct, and then an improved necessary and sufficient condition is presented.

• Journal of Magnetics
• /
• v.22 no.1
• /
• pp.122-132
• /
• 2017

This study dealt with the design process of the 78 kW permanent magnet synchronous generator for engines. After the calculation of the basic dimensions through a theoretical method in the process of initial model design, FEA (finite-element analysis) and a d,q-axis equivalent circuit were used to identify the generator characteristics depending on the number of poles. With the use of the space harmonic analysis method, the back-EMF (electromotive force) and THD were checked, and then the number of slots was determined. In addition, the most optimized generator dimensions were determined through a sizing optimization technique. Based on this, the optimum model with enhanced efficiency, material costs, and temperature characteristics was derived, and the availability of the design method was confirmed through a comparative analysis of the initial and optimum models.

• Bulletin of the Korean Space Science Society
• /
• 2010.04a
• /
• pp.35.1-35.1
• /
• 2010

Using the spectroscopic analysis and the theoretical modeling, physical properties of the nearby astrometric binary $\mu$ Cas have been determined. In spite of the well-defined parallax and astrometric orbit, there has been a chronic mass ratio problem between components. Recently, the radius of the primary component has been detected from the optical interferometric observation of the CHARA array. Using the high resolution spectroscopic analysis, we found that $\mu$ Cas have $\alpha$-enhanced chemical composition with respect to the scaled solar abundance by a factor of two. Combining our abundance analysis with recently determined physical properties, the consistent models for $\mu$ Cas have been constructed within the frame work of standard stellar theory. Through a statistical minimization between theoretical model grids, a reliable set of physical dimensions has been defined. Furthermore, the mode oscillation frequency of the best model has been calculated.

• Journal of Information Processing Systems
• /
• v.7 no.4
• /
• pp.635-652
• /
• 2011

In ubiquitous environments, many applications need to process data with time and space dimensions. Because of this, there is growing attention not only on gathering spatiotemporal data in ubiquitous environments, but also on processing such data in databases. In order to obtain the full benefits from spatiotemporal data, we need a data model that naturally expresses the properties of spatiotemporal data. In this paper, we introduce three spatiotemporal data models extended from temporal data models. The main goal of this paper is to determine which data model is less complex in the spatiotemporal context. To this end, we compare their query languages in the complexity aspect because the complexity of a query language is tightly coupled with its underlying data model. Throughout our investigations, we show that it is important to intertwine space and time dimensions and keep one-to-one correspondence between an object in the real world and a tuple in a database in order to naturally express queries in ubiquitous applications.

• Journal of the Chungcheong Mathematical Society
• /
• v.22 no.2
• /
• pp.245-250
• /
• 2009

We consider a code function from the unit interval which has a generalized dyadic expansion into a coding space which has an associated ultra metric. The code function is not a bi-Lipschitz map but a dimension-preserving map in the sense that the Hausdorff and packing dimensions of any subset in the unit interval and its image under the code function coincide respectively.