• Journal of the Korean Data and Information Science Society
• /
• v.16 no.4
• /
• pp.717-724
• /
• 2005

We often extract a new feature from the original features for the purpose of reducing the dimensions of feature space and better classification. In this paper, we show feature extraction method based on independent component analysis can be used for classification. Entropy and mutual information are used for the selection of ordered features. Performance of classification based on independent component analysis is compared with principal component analysis for three real data sets.

• Communications for Statistical Applications and Methods
• /
• v.20 no.3
• /
• pp.175-184
• /
• 2013

Kernel principal component analysis(PCA) maps observations in nonlinear feature space to a reduced dimensional plane of principal components. We do not need to specify the feature space explicitly because the procedure uses the kernel trick. In this paper, we propose a graphical scheme to represent variables in the kernel principal component analysis. In addition, we propose an index for individual variables to measure the importance in the principal component plane.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.1
• /
• pp.141-147
• /
• 2006

BOK-X12-ARIMA has been developed by the Bank of Korea in order to accomodate special features such as lunar effect, labor day and election effect which are intrinsic in Korean seasonal time series. Irregular component resulting from BOK-X12-ARIMA is usually treated as white noise time series. If this shows dependency structure, it may be advisable to incorporate dependency in irregular component into prediction. This article illustrates how to refine point forecast using dependency structure in irregular component.

• Proceedings of the ESK Conference
• /
• 1996.04a
• /
• pp.276-289
• /
• 1996

What kind of the mini-component the X-generation wanted\ulcorner It is not easy to find a mini-component which everbody can satisfy with a design and functions. By pre-test of the survey the most wanted things in the mini-component was extracted. By the post-test of the survey the newly design with a price of 500, 000 won. In this paper we try to suggest a mini-component whic X-generation can satisfy with a design and functions. Therefore, this new designed mini-component can give a visual happiness and enjoyment of the sound.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.496-498
• /
• 2003

Signal is composed of the independent components that can describe itself. These components can distinguish itself from any other signals and be extracted by analysis itself. This algorithm is called Independent Component Analysis (ICA) and image signal is considered as linear combination of independent components and features that is the weighted vector of independent component. This algorithm is already used in order to extract the good feature for image classification and very effective In this paper, we'll explain the method of stereo matching using independent component analysis and show the experimental result.

• Proceedings of the Korean Magnestics Society Conference
• /
• 2007.12a
• /
• pp.135-135
• /
• 2007

• Kyungpook Mathematical Journal
• /
• v.48 no.2
• /
• pp.241-254
• /
• 2008

We define the crosscap number of a 2-component link as the minimum of the first Betti numbers of connected, non-orientable surfaces bounding the link. We discuss some properties of the crosscap numbers of 2-component links.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.34 no.2
• /
• pp.144-158
• /
• 1998

In order to study basic information on the developed electro-magnetic compass, experiments were carried out on board M. S. KAYA at the pier of Dong Kuk Steel Mill in Pusan and the Korean southern sea using a three-axis magnetic sensor from Jan. 21, 1995 to Feb. 14, 1996. The obtained results were as follows : 1. The amount of old metal on the pier was about 27,290tons~57,440tons with an average of 40,560tons, the artificial local disturbance at the pier was min. 27.1$\mu$T, max. 66.5$\mu$T, ave. 433$\mu$T for the horizontal component and min. -27.0$\mu$T, max. 45.1$\mu$T, ave. 3.7$\mu$T for the vertical component. Its direction of horizontal component was 305$^{\circ}$ with the ship's head up bearing at 225$^{\circ}$. 2. The ship's magnetic distribution on the starboard side on berthing at the pier was 17.4$\mu$T for the horizontal component and -6.2$\mu$T for the vertical component. On the ship's port side, it was 19.8$\mu$T for the horizontal component and 4.1$\mu$T for the vertical component. On the ship's starboard side at sea, the ship's magnetic distribution was 19.2$\mu$T for the horizontal component and 3.2$\mu$T for the vertical component. On the ship's port side, the readings were 22.0$\mu$T for the horizontal component and -1.8$\mu$T for the vertical component. The directions of these readings were nearly starboard side. 3. On the pier, the secular change of the artificial local disturbance decreased 8.3$\mu$T from 61.0$\mu$T to 52.7$\mu$T for the horizontal component and decreased 7.1$\mu$T from 8.9$\mu$T M 1.8$\mu$T for the vertical component. On the starboard side from its berth, the ship, s magnetic distribution increased 2.6$\mu$T from 14.8$\mu$T to 17.4$\mu$T for the horizontal component and increased -0.1$\mu$T from -6.1$\mu$T to -6.2$\mu$T for the vertical component. On the ship's port side from its berth, it increased 7.1$\mu$T from 12.7$\mu$T to 19.8$\mu$T for the horizontal component and increased 10.2$\mu$T from -6.1$\mu$T to 4.1$\mu$T for the vertical component. 4. While at sea, on the ship's starboard side, the Secular change of the ship's magnetic distribution increased 3.9$\mu$T from 15.3$\mu$T to 19.2$\mu$T for the horizontal component and increased 2.0$\mu$T from -5.2$\mu$T to -3.2$\mu$T for the vertical component. On the port side, the changes increased 11.4$\mu$T from 10.6$\mu$T to 22.0$\mu$T for the horizontal component and increased 4.9$\mu$T from -6.7$\mu$T to -1.8$\mu$T for the vertical component. Upon berthing at the pier, the deviation of the secular change increased westerly 1 degree W~ 2.5$^{\circ}$ W from 3.5$^{\circ}$ W~ 5$^{\circ}$ W M 6W with the ship's head up bearing at 225$^{\circ}$. While at sea, these increased westerly 2$^{\circ}$ ~ 3$^{\circ}$ from the Northeast to the South and increased easterly 1$^{\circ}$ ~ 8$^{\circ}$ from the Southwest to the North. 5. While at port, within 1 mile between the ship and berth of the pier, as we approached the pier, the westerly deviation increased and when we departed the pier easterly deviation increased. When approaching the pier, the deviation was smaller than the deviation when the ship was departing from the pier. When approaching the bearing at 225$^{\circ}$ with the ship's head up bearing, the varies of deviation was smaller than the varies when the ship's head up bearing was departing from it.

• Journal of the Korea Society for Simulation
• /
• v.15 no.2
• /
• pp.59-69
• /
• 2006

Component-based software development is introduced as a new development paradigm in software development method. This approach is different from existing software development approach because it is based on reusable and autonomous unit, component. Therefore, component-based development(CBD)is divided into two stages; component development process and component assembly process; application development process. Component development process is the core of CBD because component has a key for good software. Currently many methodologies or tools have been introduced by various academies or industries. However, those don't suggest systematic and flexible modeling techniques adaptable easily into component development project. Existing approaches have a unique orarbitrary modeling technique or provide heuristic guidelines for component modeling. As a result, many component developers are faced with a difficult problems; how to developcomponent models, when develop which diagrams, and so on. In order to address this problem, we suggest a meta-model driven approach for component development in this paper. We provide meta-models according to both layer and development phase. We expect that suggested meta-models allow component developers to develop appropriate models of the time.

• International Journal of Fluid Machinery and Systems
• /
• v.4 no.4
• /
• pp.387-395
• /
• 2011

Single-blade centrifugal pumps are widely used as sewage pumps. However, the impeller of a single-blade pump is subjected to strong radial thrust during pump operation because of the geometrical axial asymmetry of the impeller. Therefore, to improve pump reliability, it is necessary to quantitatively understand radial thrust and elucidate the behavior and mechanism of thrust generating. This study investigates the radial thrust acting up on a single-blade centrifugal impeller by conducting experiments and CFD analysis. The results show that the fluctuating component of radial thrust increases as the flow rate deviates from the design flow rate to low or high value. Radial thrust was modeled by a combination of three components, inertia, momentum, and pressure by applying an unsteady conservation of momentum to the impeller. The sum of these components agrees with the radial thrust calculated by integrating the pressure and the shearing stress on the impeller surface. The behavior of each component was shown, and the effects of each component on radial thrust were clarified. The pressure component has the greatest effect on the time-averaged value and the fluctuating component of radial thrust. The time-averaged value of the inertia component is nearly 0, irrespective of the change in the flow rate. However, its fluctuating component has a magnitude nearly comparable with the pressure component at a low flow rate and slightly decreased with the increase in flow rate.