• The Journal of the Korea institute of electronic communication sciences
• /
• v.14 no.3
• /
• pp.547-552
• /
• 2019

Recently, the massive growth in the scale of data has been observed as a major issue in the Big Data. Furthermore, the Big Data should be preprocessed for normalization to get a high performance of the Machine learning since the Big Data is also an input of Machine Learning. The performance varies by many factors such as the scope of the columns in a Big Data or the methods of normalization preprocessing. In this paper, the various types of normalization preprocessing methods and the scopes of the Big Data columns will be applied to the SVM(: Support Vector Machine) as a Machine Learning method to get the efficient environment for the normalization preprocessing. The Machine Learning experiment has been programmed in Python and the Jupyter Notebook.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.24 no.3
• /
• pp.329-335
• /
• 2011

The finite element method (FEM) has been used for various fields like mathematics and engineering. However, the FEM has a difficulty in describing the geometric shape exactly due to its property of piecewise linear discretization. Recently, however, a so-called isogeometric analysis method that uses the non-uniform rational B-spline(NURBS) basis function has been developed. The NURBS can be used to describe the geometry exactly and play a role of basis functions for the response analysis. Nevertheless, constructing the NURBS basis functions in analysis is as costly as a meshing process in the FEM. Since the isogeometric method shares geometric data with CAD, it is possible to intactly import the model data from commercial CAD tools. In this paper, we use the Rhinoceros 3D software to create CAD models and export in the form of STEP file. The information of knot vectors and control points in the NURBS is utilized in the isogeometric analysis. Through some numerical examples, the accuracy of isogeometric method is compared with that of FEM. Also, the efficiency of the isogeometric method that includes the CAD and CAE in a unified framework is verified.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.24 no.1
• /
• pp.55-60
• /
• 2011

In the present work, isogeometric analysis in linear elasticity problem is conducted using the basis functions from NURBS. The objectives of isogeometric analysis introduced is to integrate both geometric modeling(CAD) and computational analysis(CAE), and this can be accomplished from direct usage of geometric modeling by NURBS as the computational mesh. The merit of the isogeometry analysis is that NURBS surface are able to represent exact geometry from the control points and knot vectors, and also subsequent refinement is relatively simple relatively. In order to verify the computer codes developed in this study, it has been applied to two structural models of which geometry are simple ; 1) circular cylinder subjected to the constant internal pressure loading, 2) square plate with circular hole at center subjected to uniform tension. The exact solutions of these two models are available. Convergence of the approximate solutions by the present code for the isogeometry analysis are investigated by mesh refinement with inserting knots (h-refinement) and by mesh refinement with order elevation of the basis functions (p-refinement).

• Geophysics and Geophysical Exploration
• /
• v.14 no.4
• /
• pp.298-304
• /
• 2011

We conducted magnetic survey using IBRV (Ice Breaker Research Vessel) ARAON of KORDI (Korea Ocean Research and Development Institute), ROV (Remotely Operated Vehicle) of Oceaneering Co. and three components vector magnetometer, at Apr., 2011 in the western slope of the caldera of TA25 seamount, the Lau Basin, the southwestern Pacific. The depth ranges of the survey area are from about 900 m to 1200 m, below sea level. For the deep sea magnetic survey, we made the nation's first small deep sea three components magnetometer of Korea. The magnetometer sensor and the data logger was attached with the upper part and lower part of ROV, respectively. ROV followed the planning tracks at 25 ~ 30 m above seafloor using the altimeter and USBL (Ultra Short Base Line) of ROV. The three components magnetometer measured the X (North), Y (East) and Z (Vertical) vector components of the magnetic field of the survey area. A motion sensor provided us the data of pitch, roll, yaw of ROV for the motion correction of the magnetic data. The data of the magnetometer sensor and the motion sensor were recorded on a notebook through the optical cable of ROV and the network of ARON. The precision positions of magnetic data were merged by the post-processing of USBL data of ROV. The obtained three components magnetic data are entirely utilized by finding possible hydrothermal vents of the survey area.

• The KIPS Transactions:PartD
• /
• v.11D no.1
• /
• pp.11-22
• /
• 2004

As random access memory chip gets cheaper, it becomes affordable to realize main memory-based database systems. Consequently, reducing cache misses emerges as the most important issue in current main memory databases, in which CPU speeds have been increasing at 60% per year, compared to the memory speeds at 10% per you. In this paper, we design and implement a main-memory database system for real-time mobile GIS. Our system is composed of 5 modules： the interface manager provides the interface for PDA users； the memory data manager controls spatial and non-spatial data in main-memory using virtual memory techniques； the query manager processes spatial and non-spatial query : the index manager manages the MR-tree index for spatial data and the T-tree index for non-spatial index : the GIS server interface provides the interface with disk-based GIS. The MR-tree proposed propagates node splits upward only if one of the internal nodes on the insertion path has empty space. Thus, the internal nodes of the MR-tree are almost 100% full. Our experimental study shows that the two-dimensional MR-tree performs search up to 2.4 times faster than the ordinary R-tree. To use virtual memory techniques, the memory data manager uses page tables for spatial data, non- spatial data, T-tree and MR-tree. And, it uses indirect addressing techniques for fast reloading from disk.