• Proceedings of the Korean Society of Computer Information Conference
• /
• 2015.07a
• /
• pp.179-180
• /
• 2015

에이전트 기반 마이크로 시뮬레이션에서 많이 사용하는 단순 전이확률(transition probability) 행렬이나 추정된 전이확률함수는 단순화하는 과정에서 정보 손실이 발생하고 복잡한 모델에서 사용할 수 없고 전이확률이 시간에 따라 변화하면 시간 변화를 따르는 별도의 추정이 필요로 한다. 본 연구는 이런 기존 방법의 한계를 해결하기 위하여, 다차원 전이확률표들을 이용해서 이벤트 발생을 결정함으로써 정보 손실을 줄이고 단순 행렬이나 함수로 표현하기 어려운 경우에도 이벤트 발생을 재현하고, 시간 독립적인 전이확률표를 이용해서 이벤트 발생을 결정함으로써 시간 변화를 별도로 추정하지 않고 이벤트 발생을 예측할 수 있는 방법을 제안하는데 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10a
• /
• pp.34-36
• /
• 2003

지능형 웹 서비스를 효과적으로 구현하기 위해서는 다양한 사용자들이 필요로 하는 데이터를 만족스럽게 제공할 수 있는 매치 메이킹의 구현이 중요한 과제이다. 이를 위한 관련 연구로 필터링 메커니즘을 제안하고 있는 LARKS, 브로커에이전트를 이용한 InfoSlueth, RDF 그래프 매칭 연구 및 DL 기반의 매칭 방법 등이 있다. 그러나 기존 연구들은 등급 개념을 가지는 유연한 검색 결과를 제공하지 못한다는 큰 문제점을 가진다. 본 논문에서는 기존 방법들을 개선하기 위한 노력으로서. 서비스 매치 메이킹의 결과들에 등급(랭킹)을 부터 하는 시맨틱 매치 메이커를 제안한다. 본 논문에서 제안하는 시맨틱 매치 메이커는 서비스 제공자와 서비스 요청자 사이의 유연한 매칭을 제공하여 지능형 웹 서비스를 효과적으로 수행 할 수 있게 한다. 본 논문에서 제안한 방법론은 서비스 광고 및 요청을 표현하기 위한 언어로 DAML-S를 채택하였고. DAML-S의 서비스 프로파일 뿐만 아니라 프로세스 모델 온톨로지 모두를 고려하는 새로운 접근법이다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2000.11a
• /
• pp.169-177
• /
• 2000

In general, the size of index set of the emotion-based control is smaller than that of the logic-based control. And thus, by using the concept of emotion we can control the behavior's patterns of multiple persons more softly from the global viewpoint. The principle just mentioned, we think, can be applied on fille general purpose system. In this paper we presented a n : n negotiation model in the deal based on emotional agent. Through the emotional layers of the agents we tried to show that the flexible control of the negotiation process is possible especially in case of dynamic environment.

• Journal of KIBIM
• /
• v.14 no.3
• /
• pp.22-30
• /
• 2024

Recently, LLM (Large Language Model), a rapidly developing generative AI technology, is receiving much attention in the smart construction field. This study proposes a methodology for implementing an knowledge expert system by linking BIM (Building Information Modeling), which supports data hub functions in the smart construction domain with LLM. In order to effectively utilize LLM in a BIM expert system, excessive model learning costs, BIM big data processing, and hallucination problems must be solved. This study proposes an LLM-based BIM expert system architecture that considers these problems. This study focuses on the RAG (Retrieval-Augmented Generation) document generation method and search algorithm for effective BIM data retrieval, with the goal of implementing an LLM-based BIM expert system within a small GPU resource. For performance comparison and analysis, a prototype of the designed system is developed, and implications to be considered when developing an LLM-based BIM expert system are derived.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.5
• /
• pp.205-212
• /
• 2019

Scene graph is a kind of knowledge graph that represents both objects and their relationships found in a image. This paper proposes a 3D scene graph generation model for three-dimensional indoor environments. An 3D scene graph includes not only object types, their positions and attributes, but also three-dimensional spatial relationships between them, An 3D scene graph can be viewed as a prior knowledge base describing the given environment within that the agent will be deployed later. Therefore, 3D scene graphs can be used in many useful applications, such as visual question answering (VQA) and service robots. This proposed 3D scene graph generation model consists of four sub-networks: object detection network (ObjNet), attribute prediction network (AttNet), transfer network (TransNet), relationship prediction network (RelNet). Conducting several experiments with 3D simulated indoor environments provided by AI2-THOR, we confirmed that the proposed model shows high performance.

• The Journal of the Korea Contents Association
• /
• v.18 no.7
• /
• pp.322-329
• /
• 2018

Teleoperation is defined with a master device that gives control command and a slave robot in a remote site. In this field, it is common that a human operator executes and experiences teleoperation with a virtual slave, and preliminary learns dynamic characteristic and network environment from both agents. Generally, a virtual slave has neglected forward dynamics and its kinematic model has been implemented in computer graphics. This makes a operator to experience actual feelings. This paper proposes a dynamic teleoperation model in which a robotic forward model is applied. Also, a novel compensation method is proposed to reduce the numerical error problems in forward dynamics caused by low control sampling rate. Finally, its results will be compared to the teleoperation in an actual environment.

• Journal of Intelligence and Information Systems
• /
• v.8 no.1
• /
• pp.39-62
• /
• 2002

In retail industry, it has been a major focus of marketing to identify and manage loyal customers effectively. Being established as a mature distribution channel, Internet retailing has launched various one-to-one marketing efforts and enjoyed much more fruitful outcome because it is founded on digitally enabled infrastructure. As more complicated and crowded transactions are expected, Internet retailing is in need of electronically available customer management services. This research presents architectural design of Web services for loyal customer management in Internet retailing. The fundamental models of the services are based on traditional RFM analysis. The Web services provide various agents that automate complicated loyal customer management tasks. beadily available Web services are expected to easily integrate into existing applications of any electronic retailers.

• The KIPS Transactions:PartC
• /
• v.9C no.4
• /
• pp.589-596
• /
• 2002

Tn the data transmission service of the mobile host, it is needed to reduce the bandwidth and to make a reliable error recovery. there are two appropriate methods in that. One is FEC which retransmits the parity of error recovery, in case of happening of the loss after forming the parity of error recovery by grouping original data. The other is local recovery based on server which can block the implosion of transmission. The use of FEC and local recovery based on server at the same time enables the reliable multicast. This paper suggests algorithm that recovers the loss by FEC and local recovery based on server through the solution of the problems about triangle routing, tunnel convergence by dividing mobile host into two parts-the sender and the receiver. This model about the loss of algorithm tested the efficiency of performance by the homogeneous independent loss, the heterogeneous independent loss, the shared source link loss model. The suggested algorithm confirms effectiveness in the environment where there are much data and many receivers in mobile host.

• The Journal of Society for e-Business Studies
• /
• v.6 no.3
• /
• pp.63-80
• /
• 2001

The construction of the students knowledge in e-Learning systems, namely the student modeling, is a core component used to develop e-Learning systems. However, existing e-Learning systems have many problems to share the knowledge in a heterogeneous student model and a distributed knowledge base. Because the methods of the knowledge representation are different in each e-Learning systems, the accumulated knowledge cannot be used or shared without a great deal of difficulty. In order to share this knowledge, existing systems must reconstruct the knowledge bases. Consequently, we propose a new a Case Markup Language based on XML in order to overcome these problems. A distributed e-Learning systems fan have the advantage of easily sharing and managing the heterogeneous knowledge base proposed by CaseML. Moreover students can generate and share a case knowledge to use the communication protocol of agents. In this paper, we have designed and developed a CaseML by using a knowledge markup language. Furthermore, in order to construct an intelligent e-Learning systems, we have done our research based on the design and development of the intelligent agent system by using CaseML.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.1A
• /
• pp.46-54
• /
• 2005

This paper presents an effcient mobile multicast method using the user pattern. We exploit the repetitive movement pattern of mobile node to reduce the total number of experience of graft and join procedure. We defined the locality scope by a movement pattern. While the network is included in the locality scope, the network should maintain a multicast tree even when the mobile node moves to the other network. In this way, the mobile host can receive a multicast service without a delay when it moves to the network in the locality scope later. We compare our scheme with existing schemes under the total signaling cost and the service delay time by using a discrete analytical model for cost analysis. Analytical results demonstrated that the total signaling cost and service delay time was significantly reduced through our proposed scheme.