• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.701-717
• /
• 2017

The interest of Korean society and government on future demographic structures is increasing due to rapid aging. Korea's mortality rate is decreasing, but the declined gap is variable. In this study, we compare the Lee-Carter, Lee-Miller, Booth-Maindonald-Smith model and functional data model (FDM) as well as Coherent FDM using non-parametric smoothing technique. We are then examine a reasonable model for projecting on mortality declined rate trend in terms of accuracy of mortality rate by ages and life expectancy. The possibility of using non-parametric techniques for the prediction of mortality in Korea was also examined. Based on the analysis results, FDM and Coherent FDM, which uses the non-parametric technique and reflects the trend of recent data, are excellent. As a result, FDM and Coherent FDM are good fit, and predictability is also excellent assuming no significant future changes.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.25 no.4
• /
• pp.331-338
• /
• 2012

Irregularity data inspected by EM-120, an railway inspection system in Korea includes unavoidable incomplete and erratic information, so it is encountered lots of problem to analyse those data without appropriate pre-data-refining processes. In this research, for the efficient management and maintenance of railway system, characteristics and problems of the detected track irregularity data have been analyzed and efficient processing techniques were developed to solve the problems. The correlation between track irregularity and seasonal changes was conducted based on ARIMA model analysis. Finally, time series analysis was carried out by various forecasting model, such as regression, exponential smoothing and ARIMA model, to determine the appropriate optimal models for forecasting track irregularity progress.

• Journal of KIISE:Software and Applications
• /
• v.29 no.9
• /
• pp.654-668
• /
• 2002

In this paper, we present an empirical study for improving the Korean text chunking based on machine learning and feature set selection approaches. We focus on two issues: the problem of selecting feature set for Korean chunking, and the problem of alleviating the data sparseness. To select a proper feature set, we use a heuristic method of searching through the space of feature sets using the estimated performance from a machine learning algorithm as a measure of "incremental usefulness" of a particular feature set. Besides, for smoothing the data sparseness, we suggest a method of using a general part-of-speech tag set and selective lexical information under the consideration of Korean language characteristics. Experimental results showed that chunk tags and lexical information within a given context window are important features and spacing unit information is less important than others, which are independent on the machine teaming techniques. Furthermore, using the selective lexical information gives not only a smoothing effect but also the reduction of the feature space than using all of lexical information. Korean text chunking based on the memory-based learning and the decision tree learning with the selected feature space showed the performance of precision/recall of 90.99%/92.52%, and 93.39%/93.41% respectively.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.15 no.5
• /
• pp.93-103
• /
• 2005

In this paper, we propose a RST resistant robust logo embedding technique for multimedia copyright protection Geometric manipulations are challenging attacks in that they do not introduce the quality degradation very much but make the detection process very complex and difficult. Watermark embedding in the normalized image directly suffers from smoothing effect due to the interpolation during the image normalization. This can be avoided by estimating the transform parameters using an image normalization technique, instead of embedding in the normalized image. Conventional RST resistant schemes that use full frame transform suffer from the absence of effective perceptual masking methods. Thus, we adopt $8\times8$ block DCT and calculate masking using a spatio-frequency localization of the $8\times8$ block DCT coefficients. Simulation results show that the proposed algorithm is robust against various signal processing techniques, compression and geometrical manipulations.

• ETRI Journal
• /
• v.24 no.6
• /
• pp.469-472
• /
• 2002

Speaker adaptation techniques are generally used to reduce speaker differences in speech recognition. In this work, we focus on the features fitted to a linear regression-based speaker adaptation. These are obtained by feature transformation based on independent component analysis (ICA), and the feature transformation matrices are estimated from the training data and adaptation data. Since the adaptation data is not sufficient to reliably estimate the ICA-based feature transformation matrix, it is necessary to adjust the ICA-based feature transformation matrix estimated from a new speaker utterance. To cope with this problem, we propose a smoothing method through a linear interpolation between the speaker-independent (SI) feature transformation matrix and the speaker-dependent (SD) feature transformation matrix. From our experiments, we observed that the proposed method is more effective in the mismatched case. In the mismatched case, the adaptation performance is improved because the smoothed feature transformation matrix makes speaker adaptation using noisy speech more robust.

• Korean Journal of Computational Design and Engineering
• /
• v.9 no.4
• /
• pp.294-305
• /
• 2004

Highly detailed geometric models are rapidly becoming commonplace in computer graphics and other applications. These complex models, which is often represented as complex1 triangle meshes, mainly suffer from the vast memory requirement for real-time manipulation of arbitrary geometric shapes without loss of data. Various techniques have been devised to challenge these problems in views of geometric processing, not a representation scheme. This paper proposes the new mesh structure for the compact representation and the efficient handling of the highly complex models. To verify the compactness and the efficiency, the memory requirement of our representation is first investigated and compared with other existing representations. And then we analyze the time complexity of our data structure by the most critical operation, that is, the enumeration of the so-called one-ring neighborhood of a vertex. Finally, we evaluate some elementary modeling functions such as mesh smoothing, simplification, and subdivision, which is to demonstrate the effectiveness and robustness of our mesh structure in the context of the geometric modeling and processing.

• Journal of Mechanical Science and Technology
• /
• v.14 no.9
• /
• pp.1005-1012
• /
• 2000

Numerical optimization techniques combined with a three-dimensional thin-layer Navier-Stokes solver are presented to find an optimum shape of a stator blade in an axial compressor through calculations of single stage rotor-stator flow. Governing differential equations are discretized using an explicit finite difference method and solved by a multi-stage Runge-Kutta scheme. Baldwin-Lomax model is chosen to describe turbulence. A spatially-varying time-step and an implicit residual smoothing are used to accelerate convergence. A steady mixing approach is used to pass information between stator and rotor blades. For numerical optimization, searching direction is found by the steepest decent and conjugate direction methods, and the golden section method is used to determine optimum moving distance along the searching direction. The object of present optimization is to maximize efficiency. An optimum stacking line is found to design a custom-tailored 3-dimensional blade for maximum efficiency with the other parameters fixed.

• The Korean Journal of Applied Statistics
• /
• v.11 no.1
• /
• pp.193-204
• /
• 1998

It is quite that we have to estimate the derivative of the regression function. The Bezier curve, rarely known to statisticians, is very popular in computer graphics area. In this paper, we use nonparametric method via the Bezier curve, and apply this method to real data set. This method seems to be very easy to compute and can be easily applied to other smoothing techniques.

• Communications for Statistical Applications and Methods
• /
• v.3 no.3
• /
• pp.113-123
• /
• 1996

Data-dependent (adaptive) choice of asymptotically efficient score functions for rank estimators and M-estimators of regression parameters in a linear regression model with left-truncated and right-censored data are developed herein. The locally adaptive smoothing techniques of Muller and Wang (1990) and Uzunogullari and Wang (1992) provide good estimates of the hazard function h and its derivative h' from left-truncated and right-censored data. However, since we need to estimate h'/h for the asymptotically optimal choice of score functions, the naive estimator, which is just a ratio of estimated h' and h, turns out to have a few drawbacks. An altermative method to overcome these shortcomings and also to speed up the algorithms is developed. In particular, we use a subroutine of the PPR (Projection Pursuit Regression) method coded by Friedman and Stuetzle (1981) to find the nonparametric derivative of log(h) for the problem of estimating h'/h.

• Journal of Korean Institute of Industrial Engineers
• /
• v.2 no.1
• /
• pp.85-92
• /
• 1976

In this paper, the employment planning model is developed which is a decision-making model for determining the optimum employment level with respect to varying net manpower requirement for each planing period such that total cost in a planning horizon is minimized. It is constructed as a nonlinear programming model and a dynamic programming model on the basis of studies in the areas of production smoothing and manpower scheduling. Costs for a planning period are categorized into regular wage cost, hiring cost, and overtime cost. The first is a linear function. The other two cost functions are of quadratic nature. The planning horizon of this planning model is intermediate range (five years) for which a fair planning accuracy can be guaranteed. The model considers learning period for each job class. It is simple and an optimum solution can be easily obtained by direct search techniques.