• IE interfaces
• /
• v.24 no.2
• /
• pp.97-104
• /
• 2011

Research for multi-variate classification has been studied through two kinds of procedures which are feature selection and classification. Feature Selection techniques have been applied to select important features and the other one has improved classification performances through classifier applications. In general, each technique has been independently studied, however consideration of the interaction between both procedures has not been widely explored which leads to a degraded performance. In this paper, through integrating these two procedures, classification performance can be improved. The proposed model takes advantage of KBANN (Knowledge-Based Artificial Neural Network) which uses prior knowledge to learn NN (Neural Network) as training information. Each NN learns characteristics of the Feature Selection and Classification techniques as training sets. The integrated NN can be learned again to modify features appropriately and enhance classification performance. This innovative technique is called ALBNN (Algorithm Learning-Based Neural Network). The experiments' results show improved performance in various classification problems.

• Journal of Korean Institute of Industrial Engineers
• /
• v.20 no.4
• /
• pp.99-111
• /
• 1994

Toguchi method is a systematic technique for designing high quality product at low cost. There are three steps in the Toguchi method, 1)system design, 2)parameter design, and 3)tolerance design. This paper considers the tolerance design for multiple performance characteristics which is practically important. We present two tolerance design procedures : grade selection and tolerance determining procedures. In grade selection procedure a scheme is presented that minimizes the sum of the price of low-level characteristics and the expected loss due to the variations of high-level characteristics. In tolerance determining procedure we determine the tolerances of the low-level characteristics.

• Journal of the Korean Data and Information Science Society
• /
• v.16 no.4
• /
• pp.861-870
• /
• 2005

The Bayesian model selection procedures for the shape parameter of gamma distribution are proposed in order to test that the failure rate of gamma distribution is constant, increasing or decreasing. The encompassing intrinsic Bayes factor by Beger and Pericchi (1996) based on Jeffreys prior for shape parameter is used to investigate the usefulness of the proposed Bayesian model selection procedures via both real data and pseudo data.

• Communications for Statistical Applications and Methods
• /
• v.26 no.6
• /
• pp.575-582
• /
• 2019

The problem of selecting variables in the presence of outliers is considered. Variable selection and outlier detection are not separable problems because each observation affects the fitted regression equation differently and has a different influence on each variable. We suggest a simultaneous method for variable selection and outlier detection in a linear regression model. The suggested procedure uses a sequential method to detect outliers and uses all possible subset regressions for model selections. A simplified version of the procedure is also proposed to reduce the computational burden. The procedures are compared to other variable selection methods using real data sets known to contain outliers. Examples show that the proposed procedures are effective and superior to robust algorithms in selecting the best model.

• Communications for Statistical Applications and Methods
• /
• v.25 no.6
• /
• pp.591-604
• /
• 2018

The accelerated failure time (AFT) model is a linear model under the log-transformation of survival time that has been introduced as a useful alternative to the proportional hazards (PH) model. In this paper we propose variable-selection procedures of fixed effects in a parametric AFT model using penalized likelihood approaches. We use three popular penalty functions, least absolute shrinkage and selection operator (LASSO), adaptive LASSO and smoothly clipped absolute deviation (SCAD). With these procedures we can select important variables and estimate the fixed effects at the same time. The performance of the proposed method is evaluated using simulation studies, including the investigation of impact of misspecifying the assumed distribution. The proposed method is illustrated with a primary biliary cirrhosis (PBC) data set.

• Archives of Reconstructive Microsurgery
• /
• v.19 no.1
• /
• pp.68-79
• /
• 2010

Lower abdominal tissue is regarded as an ideal donor site for the breast reconstruction because it provides large skin territory and huge amount of soft tissues enough to the breast size. However it is not easy for the surgeon to reconstruct the really natural breast, and needs the learning curve with long time experience. Author represent the various reconstruction procedures for the breast using lower abdominal tissue such as muscle sparing free TRAM, DIEP, and SIEA free flaps to reinsure like breast. Indications, and selection of reconstructive methods, surgical timing, selection of donor sites and recipient vessel of these flaps were reviewed. In addition, detailed procedures, surgical tips and secondary adjuvant procedures are described for more symmetry of reconstructed breast. The muscle sparing free TRAM, DIEP, and SIEA free flaps would be enough to provide supple, huge amount of well vascularized tissue for the breast, if these flaps were selected for the appropriate indication according to patient's general condition, obesity, the opposite breast and abdominal tissue condition. Lower abdominal tissue was able to provide versatile designs with sufficient adipose tissue without compromising the integrity of abdominal wall.

• The Korean Journal of Applied Statistics
• /
• v.3 no.1
• /
• pp.79-87
• /
• 1990

The problem of selecting a nonempty subset of K(>2) normal means with unknown variances has been studies by many authors. But the comparisions of the properties and the efficiencies of the proposed subset selection procedures have not been carried out. Thus we investigate properties of the proposed procedures and compare their performances for various cases.

• IE interfaces
• /
• v.18 no.3
• /
• pp.253-267
• /
• 2005

Customer satisfaction becomes one of the important managerial concerns associated with corporate competency in current competitive environment for Internet communication service companies. Hence, it is demanding to improve a company's customer satisfaction through the total quality management perspective. In practice, engineers as well as the management hope to find major quality characteristics with Internet communication network that is closely related to customer satisfaction, consequently aiming to the raise of their company's customer satisfaction. This paper presents an empirical relationship analysis between network quality characteristics and customer satisfaction on Internet communication. Methodologically, the relationship analysis framework is based on the regression variable selection procedures. In this framework, it is implemented that; 1) iterative model building; and 2) consistent criteria application to statistical tests for selecting significant variables. A case study shows that; 1) the customer satisfaction on the network connection seems to be more closely related to the network quality characteristics compared with the customer satisfaction on the network speed; and 2) the download disconnection rate has relatively evident relationship with the customer satisfaction on the network connection.

• Journal of the Korean Data and Information Science Society
• /
• v.19 no.2
• /
• pp.621-634
• /
• 2008

This paper addresses the problem of testing whether the means in several inverse Gaussian populations with heterogeneity are equal. The analysis of reciprocals for the equality of inverse Gaussian means needs the assumption of equal scale parameters. We propose Bayesian model selection procedures for testing equality of the inverse Gaussian means under the noninformative prior without the assumption of equal scale parameters. The noninformative prior is usually improper which yields a calibration problem that makes the Bayes factor to be defined up to a multiplicative constant. So we propose the objective Bayesian model selection procedures based on the fractional Bayes factor and the intrinsic Bayes factor under the reference prior. Simulation study and real data analysis are provided.

• Journal of the Korean Statistical Society
• /
• v.8 no.2
• /
• pp.99-105
• /
• 1979

In Thoman and Bain and Schafer and Sheffield, procedures for testing the equality of the scale parameters of two Weibull populations with a common shape parameter and procedures for selecting the Weibull population with the largest scale parameter are given. We give, in this paper, a modified procedure for the above testing and selection problems, which is more powerful than those previoulsy given.