• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.5
• /
• pp.545-551
• /
• 2005

The main objective of fraud detection is to minimize costs or losses that are incurred due to fraudulent transactions. Because of the problem's nature such as highly skewed, overlapping class distribution and non-uniform misclassification costs, it is, however, practically difficult to generate a classifier that is near-optimal in terms of classification costs at a desired operating range of rejection rates. This paper defines a performance measure that reflects classifier's costs at a specific operating range and offers a cost-sensitive learning approach that enables us to train classifiers suitable for real-world credit card fraud detection by directly optimizing the performance measure with evolutionary programming. The experimental results demonstrate that the proposed approach provides an effective way of training cost-sensitive classifiers for successful fraud detection, compared to other training methods.

• Journal of the Korean Geographical Society
• /
• v.47 no.4
• /
• pp.606-624
• /
• 2012

The main objective of this paper is to provide an adequate and comprehensive review of what has been done in South Korea in the field of geospatial analysis and modeling. This review focuses on spatial data analysis and spatial statistics, spatial optimization, and geosimulation among various aspects of the field. It is recognized that geospatial analysis and modeling in South Korea got through the initial stage during the 1990s when computer and analytical cartography and GIS were introduced, moved to the growth stage during the first decade of the $21^{st}$ century when there was a surge of relevant researches, and now is heading for its maturity stage. In spatial data analysis and spatial statistics, various topics have been addressed for spatial point pattern data, areal data, geostatistical data, and spatial interaction data. In spatial optimization, modeling and applications related to facility location problems, districting problems, and routing problems have been mostly researched. Finally, in geosimulation, while most of research has focused on cellular automata, studies on agent-based model and simulation are in beginning stage. Among all these works, some have fostered methodological advances beyond simple applications of the standard techniques.

• Journal of Digital Convergence
• /
• v.11 no.4
• /
• pp.109-119
• /
• 2013

Recent lifeline with Internet and smart-phone is a new ICT-based revolution that has transformed the structure of economics and business activities in digital economy. This paper deals with some economic perspectives on the conceptual concerns on transaction costs and its applications to energy and environments, and glocalization strategies of e-business. First, we consider substantial differences between offline and online channel transactions and categorize three different aspects of transaction costs in the purchasing process in both channels. Second, we examine the impacts of e-business that affect sustainable environments and provide some considerations on energy savings, product-process design, and logistics. Finally, we present some opportunities and challenges posted by global e-business and outline the economics of networking in achieving enhanced performance and competitive advantage through glocalization strategies of e-business.

• Korean journal of communication and information
• /
• v.74
• /
• pp.72-108
• /
• 2015

The more mobile technology evolves, the more users are trapped in mobile technology by being encouraging the replacement with state-of-the-art technology. However, the use of device is not entirely determined by technology's attributes itself. The meaning of smartness can be varied by 'how users accept and perceive immediate spatial perception from reality to mobile space' without recognizing the boundaries between them. This study focuses on the analysis of 'smart usage' for smartphone and this is verified the concept of 'smart usage' by the second-order measurement model. The result show that the concept organization of 'smart usage' had been differentiated and elicited by the six factors - 'multifunctional use readiness', 'administrative efficiency', 'embedded media', 'device connectivity', 'user-friendly optimization', and 'external connectivity(being connected). According to the conceptual factors, 'smart usage' can be explained in an individual's autonomous ability to control a mobile interface and to utilize a wide range of applications of smartphones.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.6
• /
• pp.1447-1456
• /
• 2017

Accurate inference of parameters of a stochastic rainfall generation model is essential to improve the applicability of the rainfall generation model which modeled the rainfall process and the structure of rainfall events. In this study, the model parameters of a stochastic rainfall generation model, NSRPM (Neyman-Scott rectangular pulse model), were estimated using DFP (Davidon-Fletcher-Powell), GA (genetic algorithm), Nelder-Mead, and DE (differential evolution) methods. Summer season hourly rainfall data of 20 rainfall observation sites within the Nakdong river basin from 1973 to 2017 were used to estimate parameters and the regional applicability of inference methods were analyzed. Overall results demonstrated that DE and Nelder-Mead methods generate better results than that of DFP and GA methods.

• Proceedings of the KIEE Conference
• /
• 1999.07b
• /
• pp.894-896
• /
• 1999

PID 제어는 50년의 역사를 갖기 때문에 현장의 사용자는 이 제어방식에 익숙해져 있으며, 제어장치의 구성이 간단하며 제어기의 최적동조가 가능하므로 많은 분야에서 사용되고 있다[1]. 그러나 PID 제어기에 의해서 얻은 결과에 대하여 만족하기 위해서는 많은 시행착오를 겪어야 한다. 또한 만족하는 결과를 얻었다고 할지라도 외란, 플랜트의 동특성이 바뀌는 경우 시스템을 추종하지 못하기 때문에 파라미터를 재조정하여야 한다. 유전 알고리즘은 자연세계의 진화 현상에 기초한 계산모델로서 John Holland에 의해서 1975년에 개발된 전역적인 최적화 알고리즘이며[1][2], 비선형 고차원, 불연속, 다중모드, 노이즈 함수 등에 대하여 강건함을 보여주고, 복잡한 탐색 공간에서 최적 값을 스스로 발견하는 학습 능력을 갖는다. 이 방법은 재생산, 교배, 돌연변이를 통하여 최적해를 찾은 방법으로 1989년에 D. E. Goldgerg에 의해서 체계적으로 정리된 후 여러 분야에서 응용되고 있다[3][4]. 그러나 유전 알고리즘은 목적함수만을 이용하여 해집단을 탐색하기 때문에 숙련운전자가 원하는 제어 특성 명세인 상승시간, 정착시간, 초과량(oveshoot) 둥을 구체적으로 명시하여 제어에 반영할 수 없다. 또한, 유전 알고리즘은 입력 값이 크게 바뀔 경우 다른 시스템으로 인식하여 새로운 탐색을 수행하는 단점을 가지고 있다. 본 논문은 첫째, 기준모델을 도입하여 플랜트의 성능을 기준모델로 표현하여 플랜트가 요구하는 성능지표를 정량적으로 규정하는 것이 가능하였다. 또한, 이것은 미지 플랜트 동특성을 식별하기 위한 신호로 사용되어, 플랜트의 정보를 얻는데 이용되었다. 즉, 기준모델과 플랜트 출력사이의 추종 오차 정보가 적응기구인 PID 유전제어기의 입력으로 사용되며, 구형파 입력의 경우에도 기준모델과 플랜트의 출력차는 크게 변하지 않는다. 따라서, 유전 알고리즘의 목적함수에 기준 모델을 제안 적용하여 안정적이고, 세밀한 제어를 수행하였다. 둘째, PID의 간단하면서 확실한 제어가 가능하다는 점과 전역적인 최적값을 찾을 수 있는 유전 알고리즘을 적용하여 고속제어를 요하는 직류 서보 모터(DC Servo Motor) 운전 시 실시간 파라미터 동조에 적용하였다.

• Smart Media Journal
• /
• v.3 no.2
• /
• pp.20-28
• /
• 2014

The issues of generation, delivery, and processing of information which have been treated importantly in information design field have evolved along with the evolution of the humankind. In the modern society, the vast amount of, complex, and artificial forms of information such as big-data is accounted for the majority and claims of interest focusing on how to effectively design those kinds of information are being increased. This study explored the visualization methods applied with the natural objects as motives as one of the ways for users to easily get their perception and cognition to the information. Nature has long influenced on the human figural activities. The natural objects take the optimum visual shapes and provide the diverse inspiration and emotion to the designers in the various design fields such as product design, architecture design, and so on. Through the literature studies, we suggested the compositional principles of natural objects and the principles for observing and analysing natural objects as a principle to use the natural objects for information design domain. We, also, suggested the information design approach model which is inspired the natural objects by linking those two kinds of principles to the information design's visual realization factors and explored the possibilities of utilizing of the approach model by the case studies.

• Journal of Korea Multimedia Society
• /
• v.12 no.10
• /
• pp.1418-1426
• /
• 2009

There has been a lot of interest in an effective method for background subtraction in an effort to separate foreground objects from a predefined background image. Promising results on background subtraction using statistical methods have recently been reported are robust enough to operate in dynamic environments, but generally require very large computational resources and still have difficulty in obtaining clear segmentation of objects. We use a simple running-average method to model a gradually changing background, instead of using a complicated statistical technique. We employ a single global threshold vector, optimized by a genetic algorithm, instead of pixel-by-pixel thresholds. A new fitness function is defined and trained to evaluate segmentation result. The system has been implemented on a PC with a webcam, and experimental results on real images show that the new method outperforms an existing method based on a mixture of Gaussian.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.11 no.6
• /
• pp.684-690
• /
• 2018

In society of the big data is being recognized as one of the core technologies witch is analysis of the collected information, the intelligent evolution of society seems to be more oriented society through an optimized value creation based on a prediction technique. If we take advantage of technologies based on big data about various data and a large amount of data generated during system operation, it will be possible to support stable operation and prevention of faults and failures. In this paper, we suggested an environment using the collection and analysis of big data, and proposed an derive time series prediction model for predicting failure through server performance monitoring for data collected and analyzed. It can be capable of supporting stable operation of the IT systems through failure prediction model for the server operator.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.12
• /
• pp.85-98
• /
• 1998

This paper proposes a new design method of neuro-FLC by the Lamarckian co-adaptation scheme that incorporates the backpropagation learning into the GA evolution in an attempt to find optimal design parameters (fuzzy rule base and membership functions) of application-specific FLC. The design parameters are determined by evolution and learning in a way that the evolution performs the global search and makes inter-FLC parameter adjustments in order to obtain both the optimal rule base having high covering value and small number of useful fuzzy rules and the optimal membership functions having small approximation error and good control performance while the learning performs the local search and makes intra-FLC parameter adjustments by interacting each FLC with its environment. The proposed co-adaptive design method produces better approximation ability because it includes the backpropagation learning in every generation of GA evolution, shows better control performance because the used COG defuzzifier computes the crisp value accurately, and requires small workspace because the optimization procedure of fuzzy rule base and membership functions is performed concurrently by an integrated fitness function on the same fuzzy partition. Simulation results show that the Lamarckian co-adapted FLC produces the most superior one among the differently generated FLCs in all aspects such as the number of fuzzy rules, the approximation ability, and the control performance.