• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.26 no.1
• /
• pp.93-107
• /
• 2016

Game bot playing is one of the main risks in Massively Multi-Online Role Playing Games(MMORPG) because it damages overall game playing environment, especially the balance of the in-game economy. There have been many studies to detect game bot. However, the previous detection models require continuous maintenance efforts to train and learn the game bots' patterns whenever the game contents change. In this work, we have proposed a machine learning technique using the self-similarity property that is an intrinsic attribute in game bots and automated maintenance system. We have tested our method and implemented a system to major three commercial games in South Korea. As a result, our proposed system can detect and classify game bots with high accuracy.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.48 no.7
• /
• pp.537-546
• /
• 2020

The reliability used as a performance indicator during the development of space launch vehicle should be validated by the launch success probability, and the launch data need to be fed back for reliability management. In this paper, the launch data of space launch vehicles around the world were investigated and statistically analyzed for the success probabilities according to the launch vehicle models and maturity. The Bayesian estimation of launch success probability was reviewed and analyzed by comparing the estimated success probabilities using several prior distributions and the statistical success probability. We presented the method of generating prior distribution function and considerations for Bayesian estimation.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.40 no.11
• /
• pp.957-962
• /
• 2012

Fluid/structure interaction (FSI) analysis is necessary to predict the response of a system in which aerodynamic pressure causes deformation of the structure, and vice versa. In dealing with a nonlinear behavior of the structure, however, a simple iterative algorithm of aerodynamic analysis with structural analysis yields no accurate results since aerodynamic pressure need to be changed in accordance with the deformation of structures. In this study, we explore an efficient and accurate method for integrating FSI analysis into structural nonlinear systems. During the course of nonlinear structural analysis, loading conditions are periodically updated by aerodynamic analysis. The accuracy and efficiency of the method is demonstrated with a high-aspect-ratio flexible wing of Global Hawk.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.26 no.4
• /
• pp.223-231
• /
• 2013

This paper presents a time-domain Gauss-Newton full-waveform inversion method for the material profile reconstruction in heterogeneous semi-infinite solid media. To implement the inverse problem in a finite computational domain, perfectly-matchedlayers( PMLs) are introduced as wave-absorbing boundaries within which the domain's wave velocity profile is to be reconstructed. The inverse problem is formulated in a partial-differential-equations(PDE)-constrained optimization framework, where a least-squares misfit between measured and calculated surface responses is minimized under the constraint of PML-endowed wave equations. A Gauss-Newton-Krylov optimization algorithm is utilized to iteratively update the unknown wave velocity profile with the aid of a specialized regularization scheme. Through a series of one-dimensional examples, the solution of the Gauss-Newton inversion was close enough to the target profile, and showed superior convergence behavior with reduced wall-clock time of implementation compared to a conventional inversion using Fletcher-Reeves optimization algorithm.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.45 no.9
• /
• pp.724-733
• /
• 2017

Accidents caused by upset condition leads to fatal damage to both manned and unmanned aircraft. This paper deals with real-time detection of these aircraft upset situations to prevent further severe situations. Firstly, the difference between sensor measurement and predicted measurement from Extended Kalman filter is monitored to determine whether a target aircraft goes into an upset condition or not. In addition, repeating the time update stage of the Extended Kalman filter for a specific length of time can enable future upset situation prediction. The results of aforementioned both the approaches will build a bridge to upset prevention for future generation of manned/unmanned aircraft.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.6B
• /
• pp.697-709
• /
• 2011

For improvement of wireless Internet service quality at vehicle's moving speed, it is advised to reduce the service disruption time by reducing the handover frequency on vehicle's moving path. Particularly, it is advantageous to avoid the handover to cell whose dwell time is short or can be ignored in terms of service continuity and average throughput. This paper proposes the handover scheme that is suitable for vehicle in order to improve the wireless Internet service quality. In the proposed scheme, the handover process continues to be learned before being modeled to Discrete-Time Markov Chain (DTMC). This modeling reduces the handover frequency by preventing the handover to cell that could provide service sufficiently to passenger even when vehicle passed through the cell but there was no need to perform handover. In order to verify the proposed scheme, we observed the average number of handovers, the average RSSI and the average throughput on various moving paths that vehicle moved in the given urban environment. The experiment results confirmed that the proposed scheme was able to provide the improved wireless Internet service to vehicle that moved to some degree of consistency.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.65-71
• /
• 2008

This paper proposes a structure of haptic collaboration system over high resolution tiled-display, and proposes a QoE (quality of experience) increase scheme in integrated system. Both haptic system and tiled-display system have requirements of computational power. A haptic device is unstable if haptic rendering rate is less than 1kHz, A requirement of tiled-display systeme is frame rate of display. It requires update of 30 frame fer sec. If we use these systems independently, we can satisfy each requirements. However, if we integrate two systems, performance of entire system significantly decreases because of lack of resources, and QoE of users also decrease. In this paper, therefore, we propose a QoE guaranty scheme which selectively allocates cpu resource between display update and haptic rendering. In order to increase QoE, we set a priority for visual and haptic in order to allocate more resource on visual or haptic. If a user sensitive about touch, then proposed scheme increases haptic rendering rate by allocating more resource. Otherwise a user more sensitive about visual than haptic, proposed scheme decreases haptic rendering rate and increases tiled-display update rate. Therefore, by selectively allocating limited cpu resource, proposed scheme guaranties QoE of both haptic and visual.

• Journal of Digital Convergence
• /
• v.19 no.12
• /
• pp.347-352
• /
• 2021

In this paper, we propose a method for finding feature subsets that are effective for classification in an input dataset by using a multi-agent reinforcement learning method. In the field of machine learning, it is crucial to find features suitable for classification. A dataset may have numerous features; while some features may be effective for classification or prediction, others may have little or rather negative effects on results. In machine learning problems, feature selection for increasing classification or prediction accuracy is a critical problem. To solve this problem, we proposed a feature selection method based on reinforced learning. Each feature has one agent, which determines whether the feature is selected. After obtaining corresponding rewards for each feature that is selected, but not by the agents, the Q-value of each agent is updated by comparing the rewards. The reward comparison of the two subsets helps agents determine whether their actions were right. These processes are performed as many times as the number of episodes, and finally, features are selected. As a result of applying this method to the Wisconsin Breast Cancer, Spambase, Musk, and Colon Cancer datasets, accuracy improvements of 0.0385, 0.0904, 0.1252 and 0.2055 were shown, respectively, and finally, classification accuracies of 0.9789, 0.9311, 0.9691 and 0.9474 were achieved, respectively. It was proved that our proposed method could properly select features that were effective for classification and increase classification accuracy.

• 한국가스학회:학술대회논문집
• /
• 2003.10a
• /
• pp.120-125
• /
• 2003

본 연구는 원자력 발전소에 있는 방화벽의 케이블 관통부위에 설치된 CPFS(Cable Penetration Fire Stop)시스템 안에서 일어나는 동적열전달 현상을 3 차원으로 나타낼 수 있는 시험시뮬레이터에 사용될 수학적 모델과 수치계산 알고리즘의 개발에 관한 것이다. CPFS 내에서 일어나는 열전도 현상을 나타내는 지배방정식은 주어진 조건들 하에서 포물선형 편미분방정식(Parabolic PDE)으로 나타난다. 문제를 단순화하기 위해 열의 흐름을 두 성분으로 나누었다 즉, 케이블과 평행한 선을 따라서 일어나는 열전도와 벽면과 평행한 평면 위에서 일어나는 열전도로 나누었다. 먼저 선을 따라 일어나는 동적 열전도 현상을 나타내는 PDE를 연속과완화(SOR: Successive Over-Relaxation)를 적용하여 유한한 불연속점들에 대한 연립 상미분방정식(ODE)으로 전환했고, 그 연립방정식은 ODE Solver 를 이용하여 풀 수 있었다. 둘째로, 각 불연속 점에 위치한 평면 위에서 일어나는 열전도를 계산하기 위해서, 유한요소의 합을 근사식으로 이용하여 PDE를 ODE로 전환해서 계산하는 유한요소법(Finite Element Method)이 이용된다. 여기서 시간과 공간의 함수 T(x, y, z, t)인 온도는 각 선의 점들과 각 평면의 요소들에 대해서 일정한 시간간격으로 초기온도와 경계온도를 업데이트하여 계산을 반복한다. 이러한 일련의 계산결과를 바탕으로 CPFS 시스템 내에서의 온도분포의 동적인 변화를 해석한다. 결론적으로 관통하는 케이블이 CPFS 시스템의 온도분포에 매우 중요한 역할을 한다는 것을 알 수 있다. 시뮬레이션 결과는 CPFS 내의 온도분포를 쉽게 이해할 수 있도록 3 차원 그래픽으로 나타냈으며, 상용소프트웨어 FEMLAB 으로 계산한 결과와 비교해서 개발된 모델과 계산 알고리즘의 정당성을 보였다. 맞이하고 있음을 볼 수 있다. 국내광업이 21C 급변하는 산업환경에 적응하여 생존하기 위해서는 각종 첨단산업에서 요구하는 소량 다품종의 원료광물을 적기에 공급 할 수 있는 전문화된 기술력을 하루속히 확보해야 하며, 이를 위해 고품위의 원료광물 확보를 위한 탐사 및 개발을 적극 추진하고 가공기술의 선진화를 위해 선진국과의 기술제휴 등 자원산업 글로벌화 정책이 절실히 요구되고 있음을 알 수 있다. 또한 삶의 질을 향상시키려는 현대인의 가치관에 부합하기 위해서는 각종 소비제품의 원료를 제공하는 광업의 본래 목적 이외에도 자연환경 훼손을 최소화하며 개발 할 수밖에 없는 구조적인 어려움에 직면할 수밖에 없다. 이처럼 국내광업이 안고 있는 여러 가지 난제들을 극복하기 위해서는 업계와 정부가 합심하여 국내광업 육성의 중요성을 재인식하고 새로운 마음가짐으로 관련 정책을 수립 일관성 있게 추진해 나가야 할 것으로 보인다.의 연구 결과를 요약하면 다음과 같다. 첫째, 브랜드 이미지와 서비스 품질과의 관계에서 브랜드이미지는 서비스 품질의 선행변수가 될 수 있음을 증명하였으며 4개 요인의 이미지 중 사풍이미지를 제외한 영업 이미지, 제품 이미지, 마케팅 이미지가 서비스 품질에 영향을 미치고 있음을 알 수 있다. 둘째, 지각된 서비스 품질과 가격 수용성과의 관계에서, 서비스 품질은 최소 가격에 신뢰서비스 요인에서 정의 영향을 미치고 있으나 부가서비스, 환경서비스에서는 역의 영향을 미침을 알수 있고, 최대 가격에 있어서는 욕구서비스 요인은 정의 영향을 미치지만 부가서비스의 경우에는 역의 영향을 미치고 있음을 알 수 있다. 셋째, 서비스품질과 재 방문 의도와의 관계에 있어서 서비스품질은 재 방문 의도에 영향을 미침을 알 수 있다. 따라서 브랜드 이미지는 서비스품질의 선행변수가 될 수 있으며, 서비스품질은 가격 수용성과 재방문 의도에 영향을 미치고 있음을 알 수

• Science of Emotion and Sensibility
• /
• v.15 no.1
• /
• pp.133-140
• /
• 2012

We designed a novel individualization process for improving personal emotion recognitions in real time. The designed individualization process was performed by a neutralization algorithm of physiological signals, a subjective emotion reflection of a user updated by personal emotion rules in real time. The physiological signals such as PPG(Photoplethysmography), GSR(Galvanic skin reflex), and SKT(Skin temperature) were measured and analyzed to estimate an emotion states of users. Regulating the emotion status using by emotion rules was performed by reflecting subjective evaluations. The agreement of emotion recognition between of individualization and non-individualization method was estimated by 10 undergraduates (5 females, mean age: $22.1{\pm}2.2$) of Sangmyung University. During the emotion recognition test, 45 images were randomly presented to each participant five times. In results, the proposed individualization process showed the agreement of 71.67 % which was five times higher than when the process was not applied. Therefore, in this study, we demonstrated that the individualization process was significantly useful for customizing emotion recognitions of personal users in real time. The individualization process will be able to improve satisfactions in various emotion related applications and services in the nearer future.