• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2002.05a
• /
• pp.645-651
• /
• 2002

we propose a k-nearest neighbors(k-NN) based pattern selection method. The method tries to select the patterns that are near the decision boundary and that are correctly labeled. The simulations over synthetic data sets showed promising results: (1) By converting a non-separable problem to a separable one, the search for an optimal error tolerance parameter became unnecessary. (2) SVM training time decreased by two orders of magnitude without any loss of accuracy. (3) The redundant SVM were substantially reduced.

• Current Optics and Photonics
• /
• v.1 no.4
• /
• pp.289-294
• /
• 2017

A high-performance time-division multiplexing (TDM) -based multiuser (MU) multiple-input multipleoutput (MIMO) system for efficient indoor visible-light communication (VLC) is presented. In this work, a MIMO technique based on RGB light-emitting diodes (LEDs) with selection combining (SC) is utilized for data transmission. That is, the proposed scheme employs RGB LEDs for parallel transmission of user data and transmits MU data in predefined slots of a time frame with a simple and efficient design, to schedule the transmission times for multiple users. Simulation results demonstrate that the proposed scheme offers an approximately 6 dB gain in signal-to-noise ratio (SNR) at a bit error rate (BER) of $3{\times}10^{-5}$, as compared to conventional MU single-input single-output (SISO) systems. Moreover, a data rate of 66.7 Mbps/user at a BER of $10^{-3}$ is achieved for 10 users in indoor VLC environments.

• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.411-425
• /
• 2021

Accelerated failure time (AFT) model represents a linear relationship between the log-survival time and covariates. We are interested in the inference of covariate's effect affecting the variation of survival times in the AFT model. Thus, we need to model the variance as well as the mean of survival times. We call the resulting model mean and variance AFT (MV-AFT) model. In this paper, we propose a variable selection procedure of regression parameters of mean and variance in MV-AFT model using penalized likelihood function. For the variable selection, we study four penalty functions, i.e. least absolute shrinkage and selection operator (LASSO), adaptive lasso (ALASSO), smoothly clipped absolute deviation (SCAD) and hierarchical likelihood (HL). With this procedure we can select important covariates and estimate the regression parameters at the same time. The performance of the proposed method is evaluated using simulation studies. The proposed method is illustrated with a clinical example dataset.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.04a
• /
• pp.1307-1310
• /
• 2009

Web services provide the possibility of dynamically integrating distributed service components scattered over the Internet to fulfill sophisticated business demands. However due to today's wide variety of services offered to perform a specific task, it's essential that users are supported in the eventual selection of appropriate services. An example of web services for which selection of appropriate services will be crucial is Auto Repair Services. Selecting proper service from a variety of Auto Part Shops would be result of delivering high-quality service and minimizing Auto Repair Service customer's waiting time. Therefore, in this research to assist selecting proper service, we present Functional-Level Mediator and illustrate its usage in matching customer's and web service's goals. Five matching cases have been analyzed and results from experiment have been shown. Also, taking advantage of implementing multithreaded web services which reflects concurrent activity in the real world more naturally, we have significantly minimized customer's waiting time at Auto Repair Service.

• Earthquakes and Structures
• /
• v.16 no.1
• /
• pp.29-39
• /
• 2019

A three-step screening process is presented in this article for selection of consistent earthquake records in which number of suitable ground motions is quickly screened and reduced to a handful number. Records that remain at the end of this screening process considerably reduce the dispersion of structural responses. Then, an effective method is presented for spectral matching and modification of the selected records. Dispersion of structural responses is explored using different statistical measures for each scaling procedure. It is shown that the Uniform Design Method, presented in this study for scaling of earthquake records, results in most cases in the least dispersion measure.

• Journal of Information Technology Applications and Management
• /
• v.17 no.1
• /
• pp.141-155
• /
• 2010

We develop an intelligent information system in a multilayer electronic supply chain network. Using the internet for supply chain management (SCM) is a key interest for contemporary managers and researchers. It has been realized that the internet can facilitate SCM by making real time information available and enabling collaboration between trading partners. Here, we propose a multi-agent system to analyze the performance of the elements of a supply network based on the attributes of the information flow. Each layer consists of elements which are differentiated by their performance throughout the supply network. The proposed agents measure and record the performance flow of elements considering their web interactions for a dynamic route selection. A dynamic programming approach is applied to determine the optimal route for a customer in the end-user layer.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2000.04a
• /
• pp.65-68
• /
• 2000

The problem of selecting a portfolio is to find Un investment plan that achieves a desired return while minimizing the risk involved. One stream of algorithms are based upon mixed integer linear programming models and guarantee an integer optimal solution. But these algorithms require too much time to apply to real problems. Another stream of algorithms are fur a near optimal solution and are fast enough. But, these also have a weakness in that the solution generated can't be guaranteed to be integer values. Since it is not a trivial job to tansform the scullion into integer valued one simutaneously maintaining the quality of the solution, they are not easy to apply to real world portfolio selection. To tackle the problem more efficiently, we propose an algorithm which generates a very good integer solution in reasonable amount of time. The algorithm is tested using Korean stock market data to verify its accuracy and efficiency.

• Clinical and Experimental Reproductive Medicine
• /
• v.49 no.4
• /
• pp.225-238
• /
• 2022

The ultimate goal of human assisted reproductive technology is to achieve a healthy pregnancy and birth, ideally from the selection and transfer of a single competent embryo. Recently, techniques for efficiently evaluating the state and quality of preimplantation embryos using time-lapse imaging systems have been applied. Artificial intelligence programs based on deep learning technology and big data analysis of time-lapse monitoring system during in vitro culture of preimplantation embryos have also been rapidly developed. In addition, several molecular markers of the secretome have been successfully analyzed in spent embryo culture media, which could easily be obtained during in vitro embryo culture. It is also possible to analyze small amounts of cell-free nucleic acids, mitochondrial nucleic acids, miRNA, and long non-coding RNA derived from embryos using real-time polymerase chain reaction (PCR) or digital PCR, as well as next-generation sequencing. Various efforts are being made to use non-invasive evaluation of embryo quality (NiEEQ) to select the embryo with the best developmental competence. However, each NiEEQ method has some limitations that should be evaluated case by case. Therefore, an integrated analysis strategy fusing several NiEEQ methods should be urgently developed and confirmed by proper clinical trials.

• ETRI Journal
• /
• v.41 no.3
• /
• pp.358-370
• /
• 2019

The main aim of this study is to select the optimal set of genes from microarray cancer datasets that contribute to the prediction of specific cancer types. This study proposes the enhancement of the feature selection filter algorithm based on Joe's normalized mutual information and its use for gene selection. The proposed algorithm is implemented and evaluated on seven benchmark microarray cancer datasets, namely, central nervous system, leukemia (binary), leukemia (3 class), leukemia (4 class), lymphoma, mixed lineage leukemia, and small round blue cell tumor, using five well-known classifiers, including the naive Bayes, radial basis function network, instance-based classifier, decision-based table, and decision tree. An average increase in the prediction accuracy of 5.1% is observed on all seven datasets averaged over all five classifiers. The average reduction in training time is 2.86 seconds. The performance of the proposed method is also compared with those of three other popular mutual information-based feature selection filters, namely, information gain, gain ratio, and symmetric uncertainty. The results are impressive when all five classifiers are used on all the datasets.

• Journal of Information Technology Applications and Management
• /
• v.28 no.6
• /
• pp.97-116
• /
• 2021

Due to the decrease in the school-age population and government regulations, universities have made great efforts to secure their own competitiveness. In particular, the selection of universities with financial support based on the recent evaluation of the Ministry of Education has become a major concern enough to affect the existence of the university itself. This paper extracts three-year data from 124 major private universities nationwide, and quantitatively analyzes the variables of major universities selected as self-improvement universities, competency reinforcement universities, and universities with limited financial support. As a result of estimating the selection of self-powered universities using the ordered logit model by hierarchically inputting 12 variables, student competitiveness in the metropolitan area (1.318^**), Educational Restitution Rate (4.078^***), University operation expenditure index rate (1.088^***) values were found. Significant positive coefficient values were found in the admission enrollment rate (45.98^***) and the enrollment rate (13.25^***). As a result of analyzing the marginal effects, the increase in the rate of reduction of education costs has always been positive in the selection of self-powered universities, but it was observed that the rate of increase decreases in areas of increase of 150% or more. On the contrary, the probability of becoming a Em-powered university was negative in all sectors, but on the contrary, it was analyzed that marginal effects increased at the same time point. On the other hand, the employment rate of graduates was not able to find direct significance with the result of the selection of Self powered universities. Through this paper, it is expected that each university will analyze the possibility and shortcomings of the selection of Self powered universities in policy making, and in particular, the risk of dropout of selection for the vulnerable field can be predicted using marginal effects. It can be used as major research data for both university evaluators, university officials and students.