• Korean Journal of Artificial Intelligence
• /
• v.9 no.1
• /
• pp.21-27
• /
• 2021

In this Paper, Since the 1990s, Korea's credit card industry has steadily developed. As a result, various problems have arisen, such as careless customer information management and loans to low-credit customers. This, in turn, had a high delinquency rate across the card industry and a negative impact on the economy. Therefore, in this paper, based on Azure, we analyze and predict the delinquency and delinquency periods of credit loans according to gender, own car, property, number of children, education level, marital status, and employment status through linear regression analysis and enhanced decision tree algorithm. These predictions can consequently reduce the likelihood of reckless credit lending and issuance of credit cards, reducing the number of bad creditors and reducing the risk of banks. In addition, after classifying and dividing the customer base based on the predicted result, it can be used as a basis for reducing the risk of credit loans by developing a credit product suitable for each customer. The predicted result through Azure showed that when predicting with Linear Regression and Boosted Decision Tree algorithm, the Boosted Decision Tree algorithm made more accurate prediction. In addition, we intend to increase the accuracy of the analysis by assigning a number to each data in the future and predicting again.

• Journal of Energy Engineering
• /
• v.17 no.2
• /
• pp.77-87
• /
• 2008

Decision-making of the site for the medium and low-level radioactive-waste disposal facilities in 2005 can be estimated as a success. But the limits exposed during the process still remain as problems to be solved. Analyzing the causes of success and failure of the policy and their correlation was expected to provide an effective guideline on future policies. The analysis shows that the transparency of policy makers, the level of community supports and the public relations are decisive factors. System dynamics, a synthetic analyzing tool, is used as a methodology for policy analysis. The result of the system dynamics analysis shows that public confidence is to be the key role to for and against logics when transparency of stakeholder, subsidy and public information are set as adjustable parameters. Public confidence takes a role of leverage that can convert tendency of conclusion by the opinion which influenced by selected parameters.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.2
• /
• pp.113-118
• /
• 2011

In this paper, we develop a speaker gender classification technique using collaborative sensor fusion for use in a wireless sensor network. The distributed sensor nodes remove the unwanted input data using the BER(Band Energy Ration) based voice activity detection, process only the relevant data, and transmit the hard labeled decisions to the fusion center where a global decision fusion is carried out. This takes advantages of power consumption and network resource management. The Bayesian sensor fusion and the global weighting decision fusion methods are proposed to achieve the gender classification. As the number of the sensor nodes varies, the Bayesian sensor fusion yields the best classification accuracy using the optimal operating points of the ROC(Receiver Operating Characteristic) curves_ For the weights used in the global decision fusion, the BER and MCL(Mutual Confidence Level) are employed to effectively combined at the fusion center. The simulation results show that as the number of the sensor nodes increases, the classification accuracy was even more improved in the low SNR(Signal to Noise Ration) condition.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.19 no.1
• /
• pp.145-155
• /
• 2010

This paper presents a new approach to obstacle avoidance for mobile robot in unknown or partially unknown environments. The method combines two navigation subsystems: low level and high level. The low level subsystem takes part in the control of linear, angular velocities using a multivariable PI controller, and the nonlinear position control. The high level subsystem uses ultrasonic and IR sensors to detect the unknown obstacle include static and dynamic obstacle. This approach provides both obstacle avoidance and target-following behaviors and uses only the local information for decision making for the next action. Also, we propose a new algorithm for the identification and solution of the local minima situation during the robot's traversal using the set of fuzzy rules. The system has been successfully demonstrated by simulations and experiments.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.36 no.3
• /
• pp.107-120
• /
• 2011

The increased availability of store brand suppliers now provides retailers with opportunities to create their own lines of vertically differentiated multiple store brands within a product category. As the number of store brands increase, the retailer's shelf space becomes more crowded, which may force the retailer to consider dropping some national brands from its assortment. Despite these trends, the problem of product line design in a vertically differentiated product category has been analyzed mainly from a manufacturer's perspective in the marketing literature and it is not known to what extent the findings of the existing product line design literature provide applicable strategic guidelines for the new problem faced by retailers. In this study, we address this deficiency in the literature and conduct an in-depth study of the retailer's strategic design of a line of store brands and its assortment decision within the context of retail category management. We analyze the retailer's decision about not only how to design a line of store brands but also which national brand to drop from its assortment. The results of our analysis are as follows. First, if the retailer has to drop one of national brands from its assortment, it is the best for the retailer to drop the low-quality national brand rather than the high-quality national brand. Second, the retailer has to position the high-quality store brand relatively close to the high-quality national brand, remained on its shelf, in terms of quality so as to maximize the size of retail margin from the national brand. On the other hand, the retailer should set the quality of the low-quality store brand at a lower level than that of the low-quality national brand to increase the total category demand by attracting more price sensitive consumers. By doing so, the retailer can also minimize cannibalization between two store brands. Lastly, our analysis shows that the introduction of a line of store brands improves consumer welfare by increasing real values of all products on the shelf.

• Proceedings of the KIEE Conference
• /
• 2003.04a
• /
• pp.163-165
• /
• 2003

Over the past several years, the application extent of the real-time systems is being expanded with the progress of civilization. An effort to minimize power consumption at the system is being accomplished in several fields from the design of an analog/digital circuit up to the device level Things of this effort have included the power optimum-technique to minimize power consumption at the digital logic circuit and the dynamic managed skill by means o( the decision of the operating system. In this paper, we designed of low power system by using Power-optimized method. As an effective low-power design, we designed the low power system which it has a monitoring system within the main board and a personal computer.

• Journal of Digital Convergence
• /
• v.10 no.10
• /
• pp.539-544
• /
• 2012

Most of the man-made systems can be modeled as a hybrid system which consists of both the high-level and the low-level component model. High level model is responsible for decision-making and the low-level one takes control of the mechanical component parts. Since the two models requires different interpretation method according to their type, analysis of a hybrid system becomes a difficult job. For the Analysis of the high-level model, methods for discrete event system models such as FSM can be used. On the contrary, numerical analysis techniques are required for the low-level continuous-time system model. Since it becomes a difficult thing for a modeller specifies and develops both the two-level models altogether, we propose an efficient hybrid simulation method which employs a game physics engine that has been widely and successfully used in the area of game industry.

• Journal of Naturopathy
• /
• v.8 no.1
• /
• pp.29-39
• /
• 2019

Purposes: The purpose of this study was to investigate the satisfaction of each activity in the high school students (568 students) after school career education (Career and vocation, Curriculum-related career education, Special lecture related to career, Entrance exam session, Career psychology test, Internet lecture), and then to analyze the career decision-making self-efficacy of the items Goal selection, Job information, Problem solving and Future plan by high, middle, and low group. Methods: Questionnaires administered to subjects after career education counseling education. Results: Career decision - making self - efficacy was significantly higher in the high group than in the middle and low groups in the goal selection, job information, problem-solving and future planning by the educational satisfaction on career and vocation, curriculum-related lectures, entrance examination session (p<.001). The level of satisfaction of special career lectures was in the order of low group> high group> middle group in career decision self - efficacy, and the values of high and low groups were significantly higher than those of the middle group. The effect of satisfaction on career decision self - efficiency in Career psychological examination education was in the order of top> middle> low groups in goal choice and job information, and the high group values were significantly higher than the middle and low groups. In the problem solving and future planning, there was a difference between high, middle, and low group values, but there was no significant difference. The effect of the satisfaction of the internet course on the career decision self - efficacy was not significant, but there was a difference between the high, middle, and low groups. Satisfaction with the internet course was low and did not affect the decision - making self - efficacy significantly. Conclusion: The higher the satisfaction of career education, the higher self - determination efficacy of goal selection, job information, problem solving and future plan. This study suggests the direction of career education.

• Journal of Practical Engineering Education
• /
• v.13 no.3
• /
• pp.473-482
• /
• 2021

This study attempted to provide basic data for career counseling and education programs by verifying the mediating effect of career decision-making self-efficacy on relationship between perceived social support and career preparation behavior of Korean Polytechnics students. A total of 314 students from 11 departments attending the I Campus of Korea Polytechnics II were surveyed, and the research results are as follows. First, it was found that career decision-making self-efficacy completely mediated the relationship between social support and career preparation behavior. Second, the career decision-making self-efficacy had complete mediating effects on the relation between social support and career preparation behavior. Third, it was found that social support indirectly had a positive effect on career preparation behavior through career decision-making self-efficacy. The results of this study show that perceived social support of Korean Polytechnics students affects career decision-making self-efficacy and increases the level of career preparation behavior. This suggests that through active social support around Korean Polytechnics students, it is necessary to provide environmental factors that can promote career preparation behavior by enhancing the self-efficacy of career decision-making, which is a low-level factor for students.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 1994.04a
• /
• pp.55-67
• /
• 1994

Studies on the ship's automatic navigation & berthing control have been continued by way of solving the ship's mathematical model but the results of such studies have not reached to our satisfactory level due to its non-linear characteristics ar low speed. In this paper the authors propose a new berthing control system which can evaluate as closely as captain's decision-making by using the FNN(Fuzzy Neural Network) controller which can simulate captain's decision-making by using the FNN(Fuzzy neural Network) controller which can simulate captain's knowledge. This berthing controller consists of the navigation subsystem FNN controller and the berthing subsystem FNN controller. The learning data are drawn from Ship Handling Simulator (NavSim NMS90 MK III) and represent the ship motion characteristics internally According to learning procedure both FNN controllers can tune membership functions and identify fuzzy control rules automatically The verified results show the FNN controllers effective to incorporate captain's knowledge and experience of berthing.