• 경영정보학연구
• /
• 제22권1호
• /
• pp.59-72
• /
• 2020

인공지능이 다루기 어려운 개념에서 아주 익숙한 도구로 자리매김 하고 있다. 이와 더불어 금융권에서도 인공지능 기술을 도입하여 기존 시스템의 문제점을 개선하고자 하는 추세이며, 그 대표적인 예가 이상거래탐지 시스템(Fraud Detection System, FDS)이다. 결제 수단의 다양화 및 전자금융거래의 증가에 따라 치밀해져 가는 사이버 금융사기(Fraud)를 기존의 규칙기반 FDS로는 탐지하기 어려워지고 있다. 이를 극복하기 위해 딥러닝 기술을 적용하여 이상거래 탐지율을 향상시키고, 이상행위에 즉각 대응하며, 탐지 결과의 반영을 자동화하고자 하는 시도가 이루어지고 있다. 딥러닝 FDS 구축에서 핵심 문제는 데이터 불균형과 이상거래 패턴의 변동이다. 본 논문에서는 생성적 적대 신경망(Generative Adversarial Network, GAN)을 활용한 오버샘플링 기법을 통해 데이터 불균형 문제를 개선하고, 이상거래 분류기로써 심층 신경망(Deep Neural Network, DNN)과 합성곱 신경망(Convolutional Neural Network, CNN)을 적용하여 이러한 문제를 개선하고자 하였다. 실험 결과, GAN 오버샘플링이 이상거래 데이터의 불균형 문제를 개선하는데 효과를 보였으며, WGAN이 가장 높은 개선 효과가 있음을 확인하였다. 또한 제안 FDS 모형의 AUC가 0.9857로 랜덤포레스트 FDS 모형에 비해 약 6.5% 향상되어, 딥러닝이 이상거래 탐지에 뛰어난 성능을 가짐을 입증하였다. 더불어 딥러닝 모형 중 DNN은 CNN에 비해 오버샘플링의 효과를 더 잘 반영함을 확인하였다.

• 한국CDE학회논문집
• /
• 제20권3호
• /
• pp.298-311
• /
• 2015

Nowadays, automatic digging operation of an excavator is a big challenge due to the complexity of digging environment, the hardness of soil and buried obstacles into the ground. In order to achieve the maximum soil bucket volume, this paper introduces a novel engineering model that was developed as a virtual excavator in the design phase. Through this model, the designs of mechanical and control systems for autonomous excavator are executed and modified easily before developing in real testbed. Based on a concept of an autonomous excavation, a mechanical system of excavator was first designed in SOLIDWORKS, and a soil model also was modeled by finite-element analysis in ANSYS, both modeled models were then exported to ADAMS environment to investigate the digging behavior through virtual simulation. An intelligent control strategy was generated in MATLAB/Simulink to control the excavator operation. The simulation results were demonstrated by effectiveness of the proposed excavator robot in testing scenarios with many soil types and obstacles.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 The Third Asian Fuzzy Systems Symposium
• /
• pp.240-245
• /
• 1998

Fuzzy information retrieval which uses the concept of fuzzy relation is able to retrieve documents in the way based on not morphology but semantics, dissimilar to traditional information retrieval theories. Fuzzy information retrieval logically consists of three sets : the set of documents, the set of terms and the set of queries. It maintains a fuzzy relational matrix which describes the relationship between documents and terms and creates a thesaurus with fuzzy relational product. It also provides the user with documents which are relevant to his query. However, there are some problems on building a thesaurus with fuzzy relational product such that it has big time complexity and it uses fuzzy values to be processed with flating-point. Actually, fuzzy values have to be expressed and processed with floating-point. However, floating-point operations have complex logics and make the system be slow. If it is possible to exchange fuzzy values with binary values, we could expect sp eding up building the thesaurus. In addition, binary value expressions require just a bit of memory space, but floating -point expression needs couple of bytes. In this study, we suggest a new method of building a thesaurus, which accelerates the operation of the system by pre-applying an ${\alpha}$-cut. The experiments show the improvement of performance and reliability of the system.

• 대한수학교육학회지:수학교육학연구
• /
• 제23권2호
• /
• pp.117-133
• /
• 2013

수학적 오개념은 학습 부진 학생들이 올바른 수학 학습을 하는 데 큰 방해 요소 중 하나이다. 오개념의 발생을 줄이고 올바른 개념 형성을 돕고자 중학교 수학 학습에서 흔히 발생하는 수학적 오개념을 정리하여 '중학교 수학과 학습 부진 학생 지도를 위한 맞춤형 교수 학습 방법 자료'를 개발하였다. 이는 수학과 학습 부진 학생 지도를 담당한 교사들에게 교수 학습에 필요한 주요 교육 내용과 창의적이고 흥미 있는 수업 아이디어를 제공하기 위한 자료이다. 뿐만 아니라 오개념을 가진 학생들이나 학습부진아 학생들의 수업자료로 사용하거나, 교사들의 연수 자료로도 사용할 수 있을 것이다.

• 한국IT서비스학회지
• /
• 제10권4호
• /
• pp.259-268
• /
• 2011

A robot education system by emulation based on Web can be efficiently used for understanding concept of robot assembly practice and control mechanism of robot by control programming. It is important to predict the path of the line tracer robot which has to be decided by the robot. Shortest Path Algorithm is a well known algorithm which searches the most efficient path between the start node and the end node. There are two related typical algorithms. Dijkstra Algorithm searches the shortest path tree from a node to the rest of the other nodes. $A^*$ Algorithm searches the shortest paths among all nodes. The delay time caused by turning the direction of navigation for the line tracer robot at the crossroads can give big differences to the travel time of the robot. So we need an efficient path determine algorithm which can solve this problem. Thus, It is necessary to analyze the overhead of changing direction of robot at multi-linked node to determine the next direction for efficient routings. In this paper, we reflect the real delay time of directional changing from the real robot. A speed based Dijkstra algorithm is proposed and compared with the previous ones to analyze the performance.

• 전기학회논문지
• /
• 제59권6호
• /
• pp.1041-1045
• /
• 2010

Short-term electric load forecasting of power systems is essential for the power system stability and the efficient power system operation. An accurate load forecasting scheme improves the power system security and saves some economic losses in power system operations. Due to scarcity of the historical same type of holiday load data, most big electric load forecasting errors occur on load forecasting for the holidays. The fuzzy linear regression model has showed good accuracy for the load forecasting of the holidays. However, it is not good enough to forecast the load of the election day. The concept of the load variation rate for the load forecasting of the election day is introduced. The proposed algorithm shows its good accuracy in that the average percentage error for the short-term 24 hourly loads forecasting of the election days is 2.27%. The accuracy of the proposed 24 hourly loads forecasting of the election days is compared with the fuzzy linear regression method. The proposed method gives much better forecasting accuracy with overall average error of 2.27%, which improved about average error of 2% as compared to the fuzzy linear regression method.

• 한국실내디자인학회논문집
• /
• 제22권3호
• /
• pp.127-134
• /
• 2013

Baroque is fundamentally in contrast with renaissance style classicism, which attempted to fix a world view itself as the value of perpetuity. Furthermore, it is interpreted as a topic of incompletion or transformation that is not defined or is not exemplary and which may be enjoyed by people in the classical life, which requires a microscopic concept in contrast with dualism, the central axis of western civilization, namely, rational lucidity. Also, such characteristics are estimated to serve as a lubricant in life while going beyond the formal aesthetic level of art even from a contemporary perspective, and to play a productive role and to have potential. Consequently, this paper assumed that the concepts expressed by the 20th century US contemporary architect Richard Meier of a complex space, namely architectural formative vocabularies such as overlapping, interpenetration, collision, and collage style mixture created by a layered structure, in his architecture are in common with the double surface technique and juxtaposition of a flat structure of Baroque architecture. Then, the paper analyzed and considered the characteristics. The paper considered the following two issues: Such Baroque style formal principle is reinterpreted against a backdrop in which his architectural flow based on the domino theory of Le Corbusier became established as his own unique architectural feature, and it may be proposed as one big flow that enables us to seek again modernistic architecture that attempts to make an architectural space absolute as an atypical formal contour.

• 한국자동차공학회논문집
• /
• 제20권4호
• /
• pp.76-81
• /
• 2012

In recent, there are tremendous efforts to apply co-generation concept in automobile to improve its thermal efficiency. The co-generation is basically a simple Rankine Cycle that uses the waste heat from the engine exhaust and coolant for heat source. In spite of developed nano technology and advance material science, the bulky co-generation system is still a big concern in automotive application. Therefore, the system should be effectively designed not to add much weight on the vehicle, but the capacity of the waste heat recovery should be still large. With such a goal in mind, the system thermal efficiency was investigated in terms of the system operation condition and working fluid. This paper provides a direction for the optimal design of the automotive co-generation system.

• 한국도서관정보학회지
• /
• 제38권3호
• /
• pp.455-475
• /
• 2007

최근의 도서관 변화의 가장 큰 특징으로 먼저 이용자의 변화를 들 수가 있는데 평일의 학생이용중심에서 주부, 무직자, 고령자 이용과 휴일에는 가족동반이용 등으로 이용자의 속성이 변화하고 있고 이에 따라 체재형도 서관이라고 하는 새로운 개념의 도서관이 출현하여 이용자의 속성에 대응하는 형태로 독서서비스를 제공하고 있다. 본 연구는 체재형 도서관으로의 이행시기에 있는 도심의 공공도서관을 대상으로 도서관내에서의 가족의 이용행동 조사하여 가족이용유형의 구성특징 및 공간구성과의 상관관계를 분석하고 비독서행위에 대하여 고찰을 행하였다.

• 한국컴퓨터산업학회논문지
• /
• 제6권5호
• /
• pp.821-828
• /
• 2005

This paper was proposed Intelligent and wireless USN/RFID module system that can overcome disadvantage of existing RFID system with no sensing module and wire communication. The proposed USN/RFID system was designed and fabricated. After fabricating new system, we analyzed the characteristics of USN/RFID module. After design VCO block that is point circuit to develop next generation system one chip of RFID system. we were carried out simulation and verified the validity, this paper was showed that VCO system was enough usable in wireless network module. USN/RFID Reader module shows superior result that validity awareness distance corresponds to 30M in the case of USN and to 5M in RFID Reader's case and 900MHz of commercial frequency does practical use enoughly in range of high frequency. The USN/RFID Reader module is considered to act big role to Ubiqitous industry offering computing surrounding of new concept that is intelligence type service and that was associated to real time location system(RTLS). environment improvement/supervision. national defense, traffic administration etc.