• 한국해양공학회지
• /
• 제36권5호
• /
• pp.295-302
• /
• 2022

The initial response to a marine accident can play a key role to minimize the accident. Therefore, various decision support systems have been developed using sensors, simulations, and active response equipment. In this study, we developed an algorithm to predict damage locations using ship motion data with bidirectional long short-term memory (BiLSTM), a type of recurrent neural network. To reflect the low frequency ship motion characteristics, 200 time-series data collected for 100 s were considered as input values. Heave, roll, and pitch were used as features for the prediction model. The F1-score of the BiLSTM model was 0.92; this was an improvement over the F1-score of 0.90 of a prior model. Furthermore, 53 of 75 locations of damage had an F1-score above 0.90. The model predicted the damage location with high accuracy, allowing for a quick initial response even if the ship did not have flood sensors. The model can be used as input data with high accuracy for a real-time progressive flooding simulator on board.

• 지능정보연구
• /
• 제16권1호
• /
• pp.71-92
• /
• 2010

최근 트레이딩 시스템에 대한 관심이 높아지면서, 인공지능을 이용한 지능형 트레이딩 시스템의 개발과 관련한 연구들이 활발하게 이루어지고 있다. 그러나 현재까지 소개된 트레이딩 시스템 관련 연구들은 트레이딩에 적용될 수 있는 다양한 변수들이 실무에서 활용되고 있음에도 불구하고, 주가지수에서 파생된 기술적 지표에만 과도하게 의존하는 경향이 있었다. 또한, 실제 수익창출에 초점이 맞추어진 트레이딩 시스템의 모형보다는 주가 혹은 주가지수의 등락에 대한 정확한 예측에 초점을 맞춰 모형을 개발하려고 하는 한계도 존재했다. 이에 본 연구에서는 기존 연구에서 주로 활용되어 온 기술적 지표 외에 현업에서 유용하게 활용되는 다양한 비가격 변수들을 시스템에 반영함으로서 예측 성과의 개선을 도모하는 동시에, Support Vector Machines 기반의 등락예측모형의 결과를 트레이딩 시스템의 매수, 매도, 혹은 유지의 신호로 해석할 수 있도록 설계된 새로운 형태의 지능형 트레이딩 시스템을 제안한다. 제안시스템의 유용성을 검증하기 위해, 본 연구에서는 2004년 5월부터 2009년 12월까지의 KOSPI200 주가지수에 제안모형을 적용하여 그 성과를 살펴보았다. 그 결과, 제안시스템이 수익률 관점에서 다른 비교모형들에 비해 더 우수한 성과를 도출함을 확인할 수 있었다.

• 한국건축시공학회지
• /
• 제13권4호
• /
• pp.416-423
• /
• 2013

An adjusted assessment model based on benefit-cost analysis (BCA) is proposed for evaluating the economic efficiency of automated construction technologies. In contrast to conventional BCA, the model does not compare monetary values, but the differences in benefits and costs between traditional and automated construction methods. To verify the usefulness of the model, it was applied to a real-scale building construction project that used a fully automated building construction system, and the face validity of the model was confirmed. The results indicate that the model can support decision makers in identifying valuable benefit factors and in assessing the cost effectiveness of the system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권2호
• /
• pp.726-741
• /
• 2022

The purpose of this study is to provide instructive theoretical models for art (music) education institutions especially when unpredictable risks, such as pandemics, occur again. Based on the customer behavior theory of the business-to-business-to-customer (B2B2C) platform, and in combination with the technology acceptance model (TAM) and expectation confirmation model (ECM), this study proposes an online education model from the perspective of art education. The framework is based on the three decision-making processes of the customer, and includes the product owner, content owner, and customer area. This paper highlights the factors that influence customers in making decisions when art education institutions are product owners. Regression analysis was introduced to study the factors influencing the expectation confirmation, and the overall fitting testing and six hypotheses testing of 385 effective samples were performed using the structural equation modeling (SEM). The results show that the course-design and after-service positively influenced the expectation confirmation, and the domain image positively influenced the continuance behavior. Negative emotions skipped the mediator (expectation confirmation) and directly exerted a significant negative impact on customers' willingness to continue system usage (continuance behavior). In addition, expectation confirmation positively influenced continuance behavior. The paths of detailed items comprising course-design, after-service, and negative emotion were also analyzed and discussed. In this path analysis, ordinary art learners did not believe that AI partners can play a very good auxiliary role. The findings contribute to the scope of information systems acting as an art education platform academically, and provide effective and theoretical support for the actual operation of art education institutions.

• 제어로봇시스템학회논문지
• /
• 제12권12호
• /
• pp.1215-1219
• /
• 2006

Scheduling plays an important role in the process management in terms of providing profit-maximizing operation sequence of multiple orders and estimating completion times of them. In order to takes its full potential, varying conditions should be properly reflected in computing the schedule. The adjustment of scheduling decisions has to be made frequently in response to the occurrence of variations. It is often challenging because their model has to be adjusted and their solutions have to be computed within short time period. This paper employs Model Predictive Control(MPC) principles for updating the process condition in the scheduling model. The solutions of the resulting problems considering variations are computed using parametric programming techniques. The key advantage of the proposed framework is that repetition of solving similar programming problems with decreasing dimensionis avoided and all potential schedules are obtained before the execution of the actual processes. Therefore, the proposed framework contributes to constructing a robust decision-support tool in the face of varying environment. An example is solved to illustrate the potential of the proposed framework with remarks on potential wide applications.

• 한국지능시스템학회논문지
• /
• 제20권4호
• /
• pp.541-547
• /
• 2010

일반적인 사고와는 달리 해양사고는 인명피해와 함께 심각한 자본의 손실을 초래한다. 이러한 해양 사고가 발생하는 원인을 살펴보면 대부분 항해사의 부주의나 잘못된 판단에 의한 것으로 추정되고 있기 때문에, 항해사의 의사결정을 보조하거나 대체할 수 있는 기술에 대한 요구가 있다. 이와 관련한 기술들을 소개하는 많은 연구들이 진행되어 오고 있지만, 항행중인 선박의 주변 환경(수심, 항로 폭, 어로구역, 조류, 해양사고 발생 건수, 해상 장애물 등)에 대한 고려는 거의 되어있지 않다. 그러나, 실제 항행 상황에서 선박 인근 환경 정보에 대한 고려는 중요한 요소이다. 이전 연구에서 우리는 항행 선박의 다양한 주변 환경 요소가 고려된 환경 위험도 평가에 관한 개념 모델을 소개하였다. 이 연구에서는 항행 전문가의 의견을 수렴하여 각 주변 환경 요소별 상세 설계 내용을 소개하고, 검증용 구현 결과의 분석을 통해 기존 개념 모델에 대한 타당성을 보인다.

• 한국컴퓨터정보학회논문지
• /
• 제19권11호
• /
• pp.149-157
• /
• 2014

당뇨병은 급성합병증을 예방하고 장기간의 합병증의 위험도를 감소하기 위하여 지속적인 치료와 환자 자가 관리 교육이 필요한 만성질환이다. 또한 전 세계적으로 당뇨병에 대한 유병률과 사망률이 대부분의 인구집단에서 역학적 비율에 도달하였다. 많은 연구에서 당뇨병에 대한 조기진단은 적절한 치료와 생활습관을 지키는 관리를 통하여 발병을 예방하는데 도움을 줄 수 있으며, 이를 통하여 당뇨병의 합병증을 감소시키고 생존률을 향상시킬 수 있다고 보고하고 있다. 본 연구는 PIMA Indians 당뇨 데이터에 대하여 mixture of expert 모형을 적용하여 당뇨유병환자의 여부를 분류하고, 이를 로지스틱 회귀분석, 신경망분석의 성능과 비교함으로서 그 유용성을 주장하고자 하였다. 연구결과 정확도 및 ROC 곡선, c-통계량에서 ME 모형이 다른 분류도구들에 비해서 높게 나타남을 확인할 수 있었다.

• 한국시뮬레이션학회논문지
• /
• 제30권3호
• /
• pp.1-9
• /
• 2021

전장에 있는 지휘관과 참모들은 상황을 인식하고 그 결과를 바탕으로 지휘결심을 통해 군사 활동을 수행하는데, 최근 정보기술의 발달과 함께 지휘결심을 지원하는 인공지능에 대한 요구가 증가하였다. 인공지능을 활용하기 위해서는 강화학습에 필요한 학습 data set의 식별, 수집 그리고 전처리가 필수적이다. 그러나 전술 C4I 체계에 저장된 적 data는 정확성, 적시성, 충분성 측면에서 인공지능 학습 data로 사용하기에 적절하지 않기 때문에 학습 data를 수집하고 훈련 시킬 수 있는 대안이 필요하다. 본 논문에서는 육군의 워게임 훈련 모델인 '창조 21 모델 훈련 data'를 활용하여 인공지능을 학습시키는 방법론을 제시하였다. 연구 범위는 군사결심수립과정과 연계하여 인공지능의 역할과 범위를 구체화하고, 그 역할에 맞추어 인공지능을 훈련 시키기 위해 창조 21 모델 연습 data를 활용하는 모델을 제시하였다. 공개가 제한되는 군사자료의 특성을 고려하여 가상의 sample data를 제작하였고, 공개가 제한되는 대한민국 육군의 교리는 인터넷에서 수집 가능한 미군 교리를 활용하였다.

• Asia pacific journal of information systems
• /
• 제18권2호
• /
• pp.15-40
• /
• 2008

With the explosive growth of the Internet, Internet shopping malls have become recognized as one of the major purchasing channels for consumers, as well as one of the competitive distribution channels for companies that allow them to contact with customers without intermediaries. It has motivated information systems(IS) researchers to examine the factors influencing consumer behavior and the purchase decision process in the context of Internet shopping malls. Despite the extensive research that has been conducted on the purchase decision process of consumers in online shopping malls, the results have demonstrated a need for further understanding of consumer behavior due to the unique features of virtual space and the characteristics of online consumers. Previous studies from marketing and consumer behavior domains have suggested that the concept of involvement plays an important role in explaining consumers' purchase behavior. Despite the critical role of involvement and the explosive growth of e-commerce, little research has examined the role of involvement in the Internet shopping mall context. With this motivation, this study has two research objectives. First, it introduces and tests an theoretical model capable of better explaining consumers' intention to purchase in the Internet shopping mall context. The proposed model extends and integrates existing models on purchase intention by incorporating purchase experience, innovativeness, and perceived self-control as the consumer factors, along with perceived risk, information provision, and perceived price as the Internet shopping mall factors. Second, this study examines how involvement differences may affect consumers' intention to purchase. For this purpose, two factors from involvement theory, involvement type and involvement level, are introduced into the research model as moderating variables. In order to test the proposed model, the overall approach employed was a field study using the structural equation model. We developed our data collection instrument by adopting existing validated questions wherever possible. All question items were measured with a seven-point, Likert-type scale, with anchors ranging from 'strongly disagree' to 'strongly agree.' Two IS researchers reviewed the instrument and checked its face validity. We collected empirical data for this study over a period of two weeks from subjects who had purchase experiences through Internet shopping malls. A total of 473 complete and valid responses were obtained. We carried out data analysis using a two-step methodology with AMOS 4.0. The first step in the data analysis was to establish the convergent and discriminant validity of the constructs. In the second step, we examined the structural model based on the cleansed measurement model. The empirical results partly support the proposed model and identify the moderating effect of involvement differences. Theoretical and practical implications of the study are discussed, along with its limitations.

• International Journal of Computer Science & Network Security
• /
• 제23권10호
• /
• pp.135-146
• /
• 2023

An effective educational program warrants the inclusion of an innovative construction which enhances the higher education efficacy in such a way that accelerates the achievement of desired results and reduces the risk of failures. Educational Decision Support System (EDSS) has currently been a hot topic in educational systems, facilitating the pupil result monitoring and evaluation to be performed during their development. Insufficient information systems encounter trouble and hurdles in making the sufficient advantage from EDSS owing to the deficit of accuracy, incorrect analysis study of the characteristic, and inadequate database. DMTs (Data Mining Techniques) provide helpful tools in finding the models or forms of data and are extremely useful in the decision-making process. Several researchers have participated in the research involving distributed data mining with multi-agent technology. The rapid growth of network technology and IT use has led to the widespread use of distributed databases. This article explains the available data mining technology and the distributed data mining system framework. Distributed Data Mining approach is utilized for this work so that a classifier capable of predicting the success of students in the economic domain can be constructed. This research also discusses the Intelligent Knowledge Base Distributed Data Mining framework to assess the performance of the students through a mid-term exam and final-term exam employing Multi-agent system-based educational mining techniques. Using single and ensemble-based classifiers, this study intends to investigate the factors that influence student performance in higher education and construct a classification model that can predict academic achievement. We also discussed the importance of multi-agent systems and comparative machine learning approaches in EDSS development.