• Knowledge Management Research
• /
• v.10 no.4
• /
• pp.1-15
• /
• 2009

국내에서 신용카드는 대표적인 지불 수단으로 정착되었으며 신용카드의 사용자와 신용카드의 발급 매수는 이미 포화상태에 도달해 있다. 이 같은 양적 성장은 정부의 신용카드 활성화 정책과 더불어 신용카드사 간의 과당 경쟁의 영향에 기인하고 있다. 신용차드의 사용층은 대부분의 성인 남녀로 확대되었으며, 특히 복수의 신용카드 소지자를 대상으로 자사가 발급한 신용차드를 사용하게 하기 위한 신용카드사 간의 경쟁이 치열한 상황이다. 이에 따라 신용카드사들이 경쟁사의 카드사용 회원을 자사의 회원으로 확보하는 젓이 불가피하며 마찬가지로 사용 중인 자사의 회원이 경쟁사로 이동하지 않도록 사전에 이탈 징후를 포착하여 유지 캠페인을 수행하는 것이 신용카드사 마케팅의 주요 활동이 되었다. 선행연구에서는 신용카드 회원의 이탈과 관련하여 다양한 데이터마이닝 기법을 이용한 이탈의 특성 분류 연구가 진행되었다. 본 연구는 회원 이탈에 영향을 주는 요인을 효과적으로 발견하기 위한 방법으로 베이지안 네트워크(Bayesian Network)를 활용한다. 특히, 베이지안 네트워크의 일종인 일반 베이지안 네트워크(General Bayesian Network)를 이용하여 회원의 이탈요인에 영향을 주는 요인들의 집합인 마코프 블랭킷(Makov Blanket)을 도출한다. 한편, 마코프 블랭킷에 포함된 변수를 이용해 민감도 분석을 수행하여 영향이 큰 요인을 찾아내고 이를 비즈니스 프로세스에 적용하여 실무적인 의의를 실증하고자 한다.

• Journal of the Korean Data and Information Science Society
• /
• v.12 no.2
• /
• pp.11-25
• /
• 2001

There are several algorithms for classification in modeling relations, patterns, and rules which exist in data. We learn to classify objects on the basis of instances presented to us, not by being given a set of classification rules. The Bayesian learning uses the probability distribution to express our knowledge about unknown parameters and update our knowledge by the law of probability as the evidence gathered from data. Also, the neural network models are designed for predicting an unknown category or quantity on the basis of known attributes by training. In this paper, we compare the misclassification error rates of Bayesian Neural Network method with those of other classification algorithms, CHAID, CART, and QUBST using several data sets.

• Journal of Navigation and Port Research
• /
• v.41 no.5
• /
• pp.287-296
• /
• 2017

The Formal Safety Assessment (FSA) is a structured and systematic methodology developed by the IMO, aimed at assessing the risk of vessels and recommending the method to control intolerable risks, thereby enhancing maritime safety, including protection of life, health, the marine environment and property, by using risk analysis and cost-benefit assessment. While the FSA has mostly been applied to merchant vessels, it has rarely been applied to a DP vessel, which is one of the special purpose vessels in the offshore industry. Furthermore, most of the FSA has been conducted so far by using the Fault Tree Analysis tool, even though there are many other risk analysis tools. This study carried out the FSA for safe operation of DP vessels by using the Bayesian network, under which conditional probability was examined. This study determined the frequency and severity of DP LOP incidents reported to the IMCA from 2001 to 2010, and obtained the Risk Index by applying the Bayesian network. Then, the Risk Control Options (RCOs) were identified through an expert brainstorming and DP vessel simulations. This study recommends duplication of PRS, regardless of the DP class and PRS type and DP system specific training. Finally, this study verified that the Bayesian network and DP simulator can also serve as an effective tool for FSA implementation.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.3
• /
• pp.65-75
• /
• 2019

Non-Functional Requirements (NFR) play a crucial role during the software development process. Currently, NFRs are considered more important than Functional Requirements and can determine the success of a software system. NFRs can be very complicated to understand due to their subjective manner and especially their conflicting nature. Self-adaptive systems (SAS) are operating in dynamically changing environment. Furthermore, the configuration of the SAS systems is dynamically changing according to the current systems context. This means that the configuration that manages the trade-off between NFRs in this context may not be suitable in another. This is because the NFRs satisfaction is based on a per-context basis. Therefore, one context configuration to satisfy one NFR may produce a conflict with another NFR. Furthermore, current approaches managing Non-Functional Requirements trade-off stops managing them during the system runtime which of concern. To solve this, we propose fragmentizing the NFRs and their alternative solutions in form of Multi-entity Bayesian network fragments. Consequently, when changes occur, our system creates a situation specific Bayesian network to measure the impact of the system's conditions and environmental changes on the NFRs satisfaction. Moreover, it dynamically decides which alternative solution is suitable for the current situation.

• The Journal of Korea Robotics Society
• /
• v.13 no.4
• /
• pp.213-219
• /
• 2018

In this paper, we introduce a target position reasoning system based on Bayesian network that selects destinations of robots on a map to explore compound disaster environments. Compound disaster accidents have hazardous conditions because of a low visibility and a high temperature. Before firefighters enter the environment, the robots notify information in advance, such as victim's positions, number of victims, and status of debris of building. The problem of the previous system is that the system requires a target position to operate the robots and the firefighter need to learn how to use the robot. However, selecting the target position is not easy because of the information gap between eyewitness accounts and map coordinates. In addition, learning the technique how to use the robots needs a lot of time and money. The proposed system infers the target area using Bayesian network and selects proper x, y coordinates on the map based on image processing methods of the map. To verify the proposed system, we designed three example scenarios based on eyewetinees testimonies and compared time consumption between human and the system. In addition, we evaluate the system usability by 40 subjects.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.422-422
• /
• 2022

최근 기후변화와 도시화 등의 영향으로 인해 전 지구적으로 홍수 피해의 규모와 홍수발생 빈도가 증가하고 있다. 특히, 전 세계 인구의 약 50% 이상이 거주하고 있는 연안지역의 홍수피해 위험성은 급격히 증가하고 있는 추세이며, 각 국가는 홍수 피해를 저감하고 예방하기 위한 노력을 지속적으로 기울이고 있다. 본 연구에서는 연안지역의 감조하천을 대상으로 홍수 예경보 의사결정기법을 개발하고자 하였다. 이를 위해 감조하천에서 관측된 수위는 조석에 의한 수위(조석 성분), 파고에 의한 수위(파고 성분), 강우에 의한 수위(강우-유출 성분), 그리고 잡음에 의한 수위(잡음 성분)의 4가지 수문 성분으로 구성되어 있다고 정의하였고, 감조하천의 예측 강우 성분에 해당하는 예측 수위를 추정하기 위해 수위-유량 관계 곡선식을 개발하고자 하였다. 또한 각 수문 성분별 위기 경보 단계를 설정하고, Bayesian Network를 활용하여 수문 성분들의 위험을 종합적으로 고려할 수 있는 홍수 예·경보 의사결정 기법을 개발하였다. 3가지 난수 발생 방법에 따라 Bayesian Network 모형을 통해 다양한 수문 조건에 따른 조건부 확률을 산정하였으며, 정확도 검토를 수행한 결과 F-1 Socre가 25.1%, 63.5% 및 82.3%의 정확도를 보였다. 향후 본 연구에서 제시한 방법론을 활용한다면 기상청에서 제공하고 있는 예측 강우 및 GRM 모형을 통해 유출량을 산정하고, 이를 예측 수위로 변환하여 연안 지역의 홍수 위험도 매트릭스를 통해 홍수 예·경보에 대한 의사결정을 수행할 수 있을 것으로 판단된다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.5
• /
• pp.495-503
• /
• 2014

In recent days, a study of the intrusion detection system collecting and analyzing network data, packet or logs, has been actively performed to response the network threats in computer security fields. In particular, Bayesian network has advantage of the inference functionality which can infer with only some of provided data, so studies of the intrusion system based on Bayesian network have been conducted in the prior. However, there were some limitations to calculate high detection performance because it didn't consider the problems as like complexity of the relation among network packets or continuos input data processing. Therefore, in this paper we proposed two methodologies based on K-menas clustering to improve detection rate by reforming the problems of prior models. At first, it can be improved by sophisticatedly setting interval range of nodes based on K-means clustering. And for the second, it can be improved by calculating robust CPT through applying weighted-leaning based on K-means clustering, too. We conducted the experiments to prove performance of our proposed methodologies by comparing K_WTAN_EM applied to proposed two methodologies with prior models. As the results of experiment, the detection rate of proposed model is higher about 7.78% than existing NBN(Naive Bayesian Network) IDS model, and is higher about 5.24% than TAN(Tree Augmented Bayesian Network) IDS mode and then we could prove excellence our proposing ideas.

• Journal of Internet Computing and Services
• /
• v.13 no.5
• /
• pp.21-31
• /
• 2012

Threat Evaluation(TE) which has air intelligence attained by identifying friend or foe evaluates the target's threat degree, so it provides information to Weapon Assignment(WA) step. Most of TE data are passed by sensor measured values, but existing techniques(fuzzy, bayesian network, and so on) have many weaknesses that erroneous linkages and missing data may fall into confusion in decision making. Therefore we need to efficient Threat Evaluation system that can refine various sensor data's linkages and calculate reliable threat values under unpredictable war situations. In this paper, we suggest new threat evaluation system based on information fusion JDL model, and it is principle that combine fuzzy which is favorable to refine ambiguous relationships with bayesian network useful to inference battled situation having insufficient evidence and to use learning algorithm. Finally, the system's performance by getting threat evaluation on an air defense scenario is presented.

• Journal of Korea Water Resources Association
• /
• v.48 no.10
• /
• pp.781-791
• /
• 2015

Dam risk analysis requires a systematic process to ensure that hydrologic variables (e.g. precipitation, discharge and water surface level) contribute to each other. However, the existing dam risk approach showed a limitation in assessing the interdependencies across the variables. This study aimed to develop Bayesian network based dam risk analysis model to better characterize the interdependencies. It was found that the proposed model provided advantages which would enable to better identify and understand the interdependencies and uncertainties over dam risk analysis. The proposed model also provided a scenario-based risk evaluation framework which is a function of the failure probability and the consequence. This tool would give dam manager a framework for prioritizing risks more effectively.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.4
• /
• pp.572-578
• /
• 2008

We analyze the search space considered by the previously proposed R-CORE method for learning Bayesian network structures of large scale. Experimental analysis on the search space of the method is also shown. The R-CORE method reduces the search space considered for Bayesian network structures by recursively clustering the random variables and restricting the orders between clusters. We show the R-CORE method has a similar search space with the previous method in the worst case but has a much less search space in the average case. By considering much less search space in the average case, the R-CORE method shows less tendency of overfitting in learning Bayesian network structures compared to the previous method.