• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.6 no.4
• /
• pp.277-281
• /
• 2006

Cyber counseling, one of the most compatible type of consultation for the information society, enables people to reveal their mental agonies and private problems anonymously, since it does not require face-to-face interview between a counsellor and a client. However, there are few cyber counseling centers which provide high quality and trustworthy service, although the number of cyber counseling center has highly increased. Therefore, this paper is intended to enable an appropriate consultation for each client by analyzing client propensity using Bayesian variable selection. Bayesian variable selection is superior to stepwise regression analysis method in finding out a regression model. Stepwise regression analysis method, which has been generally used to analyze individual propensity in linear regression model, is not efficient since it is hard to select a proper model for its own defects. In this paper, based on the case database of current cyber counseling centers in the web, we will analyze clients' propensities using Bayesian variable selection to enable individually target counseling and to activate cyber counseling programs.

• Journal of Korean Institute of Industrial Engineers
• /
• v.42 no.5
• /
• pp.314-326
• /
• 2016

The purpose of variable selection techniques is to select a subset of relevant variables for a particular learning algorithm in order to improve the accuracy of prediction model and improve the efficiency of the model. We conduct an empirical analysis to evaluate and compare seven well-known variable selection techniques for multiple linear regression model, which is one of the most commonly used regression model in practice. The variable selection techniques we apply are forward selection, backward elimination, stepwise selection, genetic algorithm (GA), ridge regression, lasso (Least Absolute Shrinkage and Selection Operator) and elastic net. Based on the experiment with 49 regression data sets, it is found that GA resulted in the lowest error rates while lasso most significantly reduces the number of variables. In terms of computational efficiency, forward/backward elimination and lasso requires less time than the other techniques.

• Journal of the Korean Data and Information Science Society
• /
• v.19 no.3
• /
• pp.789-800
• /
• 2008

Current credit evaluation models based on financial data make use of smoothing estimated default ratios which are transformed from each financial variable. In this work, some problems of the credit evaluation models developed by financial experts are discussed and we propose improved credit evaluation models based on the stepwise variable selection method and Box-Cox transformed data whose distribution is much skewed to the right. After comparing goodness-of-fit tests of these models, the validation of the credit evaluation models using statistical methods such as the stepwise variable selection method and Box-Cox transformation function is explained.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.50 no.4
• /
• pp.268-275
• /
• 2005

Early predictions of crop yields call provide information to producers to take advantages of opportunities into market places, to assess national food security, and to provide early food shortage warning. The objectives of this study were to identify the most useful parameters for estimating yields and to compare two model selection methods for finding the 'best' model developed by multiple linear regression. This research was conducted in two 65ha corn/soybean rotation fields located in east central South Dakota. Data used to develop models were small temporal variability information (STVI: elevation, apparent electrical conductivity $(EC_a)$, slope), large temporal variability information (LTVI : inorganic N, Olsen P, soil moisture), and remote sensing information (green, red, and NIR bands and normalized difference vegetation index (NDVI), green normalized difference vegetation index (GDVI)). Second order Akaike's Information Criterion (AICc) and Stepwise multiple regression were used to develop the best-fitting equations in each system (information groups). The models with $\Delta_i\leq2$ were selected and 22 and 37 models were selected at Moody and Brookings, respectively. Based on the results, the most useful variables to estimate corn yield were different in each field. Elevation and $EC_a$ were consistently the most useful variables in both fields and most of the systems. Model selection was different in each field. Different number of variables were selected in different fields. These results might be contributed to different landscapes and management histories of the study fields. The most common variables selected by AICc and Stepwise were different. In validation, Stepwise was slightly better than AICc at Moody and at Brookings AICc was slightly better than Stepwise. Results suggest that the Alec approach can be used to identify the most useful information and select the 'best' yield models for production fields.

• Industrial Engineering and Management Systems
• /
• v.16 no.1
• /
• pp.64-79
• /
• 2017

Identifying disease genes from human genome is a critical task in biomedical research. Important biological features to distinguish the disease genes from the non-disease genes have been mainly selected based on traditional feature selection approaches. However, the traditional feature selection approaches unnecessarily consider many unimportant biological features. As a result, although some of the existing classification techniques have been applied to disease gene identification, the prediction performance was not satisfactory. A small set of the most important biological features can enhance the accuracy of disease gene identification, as well as provide potentially useful knowledge for biologists or clinicians, who can further investigate the selected biological features as well as the potential disease genes. In this paper, we propose a new stepwise random forests (SRF) approach for biological feature selection and disease gene identification. The SRF approach consists of two stages. In the first stage, only important biological features are iteratively selected in a forward selection manner based on one-dimensional random forest regression, where the updated residual vector is considered as the current response vector. We can then determine a small set of important biological features. In the second stage, random forests classification with regard to the selected biological features is applied to identify disease genes. Our extensive experiments show that the proposed SRF approach outperforms the existing feature selection and classification techniques in terms of biological feature selection and disease gene identification.

• Korean Journal of Computational Design and Engineering
• /
• v.4 no.2
• /
• pp.100-109
• /
• 1999

The bearing design includes the steps of selection bering type, selection bearing subtype, and determining the peripheral equipments. In this paper decision making methodologies are compared to propose a stepwise decision methodology to the bearing selection problem. An artificial neural network trained with design cases is used for selecting a bearing type in the first step. Then the subtype of the bearing is selected using the weighting method, high is a kind of multi-criteria decision making method. Finally, the types of peripheral equipments such as lubrication devices, seals and bearing housings are determined using a rule-based expert system.

• Journal of Korean Society for Quality Management
• /
• v.47 no.1
• /
• pp.33-46
• /
• 2019

Purpose: This study proposed a DEA (Data Envelopment Analysis)-based stepwise benchmarking target selection for inefficient DMU (Decision Making Unit) to improve its efficiency gradually to reach most efficient frontier considering resource (DEA inputs and outputs) improvement preferences. Methods: The proposed method proceeded in two steps. First step evaluates efficiency of DMUs by using DEA, and an evaluated DMU selects benchmarking targets of HCU (Hypothesis Composit Unit) or RU (Real Unit) considering resource improvement preferences. Second step selects stepwise benchmarking targets of the inefficient DMU. To achieve this, this study developed a new DEA model, which can select a benchmarking target of an inefficient DMU in considering inputs or outputs improvement preference, and suggested an algorithm, which can select stepwise benchmarking targets of the inefficient DMU. Results: The proposed method was applied to 34 international ports for validation. In efficiency evaluation, five ports was evaluated as most efficient port, and the remaining 29 ports was evaluated as relative inefficient port. When port 34 was supposed as evaluated DMU, its can select its four stepwise benchmarking targets in assigning the preference weight to inputs (berth length, total area of pier, CFS, number of loading machine) as (0.82, 1.00, 0.41, 0.00). Conclusion: For the validation of the proposed method, it applied to the 34 major ports around the world and selected stepwise benchmarking targets for an inefficient port to improve its efficiency gradually. We can say that the proposed method enables for inefficient DMU to establish more effective and practical benchmarking strategy than the conventional DEA because it considers the resource (inputs or outputs) improvement preference in selecting benchmarking targets gradually.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.39 no.3
• /
• pp.51-61
• /
• 2014

In this paper, we deal with developing a cost estimation relationships (CER) for Korean maneuverable weapons systems using historical production cost. To develop the CER, we collected the historical data of the production cost of four tanks and five armored vehicles. We also analyzed the Required Operational Capability (ROC) of the weapons systems and chose cost drivers that can compare operational capabilities of the weapons systems We used Forward selection, Backward selection, Stepwise Regression and $R^2$ selection as the cost drivers which have the greatest influence with the dependent variables. And we used Principle Component Regression, Robust Regression and Weighted Regression to deal with multicollinearity and outlier among the data to develop a more appropriate CER. As a result, we were able to develop a production cost CER for Korean maneuverable weapons systems that have the lowest cost errors. Thus, this research is meaningful in terms of developing a CER based on Korean original cost data without foreign data and these methods will contribute to developing a Korean cost analysis program in the future.

• Korea Journal of Hospital Management
• /
• v.26 no.2
• /
• pp.17-26
• /
• 2021

Purposes: This study aims to identify factors affecting dental university hospitals' profitability and understand recent their business condition. Methodology: Data from 2016 to 2019 was collected from financial statement, public open data in 8 dental university hospitals. For the study, multiple regression test with stepwise selection was applied. Findings: First of all, 9 out of 19 independent variables were selected by stepwise selection. As a result of multiple regression test with selected independent variables and the dependent variable(operating profit margin ratio), the factors affecting hospitals' profitability were the number of dental unit chair, hospital location, debt ratio, total capital turnover ratio, employment cost rate, material cost rate, management expense rate, the number of patient per a dentist. Practical Implication: To improve dental university hospitals' profitability, hospitals specifically analysis and manage their cost such as employment, material and management cost and seek effectiveness by managing the proper number of patient per a dentist.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.39 no.2
• /
• pp.67-82
• /
• 2014

The purpose of this study is two-fold : the one is to examine the causal relationship between domestic large firms' win-win growth effort and their financial performance by fiscal years; and the other is to develop a quantitative win-win growth index to overcome the limitation of the current one mainly using a survey method developed by NCCP (National Commission for Corporate Partnership). To serve the first purpose, we take a sample of 128 large companies whose win-win growth indices as of year 2011 and 2012 were evaluated by NCCP. We use their respective fiscal year's financial data to select 62 candidate financial ratios, which are then used in subsequent empirical tests. For the tests, we employ ordered probit model with stepwise selection method and two-way ANOVA with randomized block design to identify which of the 62 financial ratios are statistically significant ones to affect the firms' win-win growth index as well as to determine if the firms' win-win growth effort would cause their financial performance positively. To serve the second purpose, we devise a model using the 123 firms' 45 financial ratios, which employs ordered probit model with stepwise selection, and the validation of the model follows. We claim that the model suggested in this study serve as an alternative complementing the current one as it can produce the index in a more objective and swift manner using the firms' publicized financial statements.