• Communications of the Korean Institute of Information Scientists and Engineers
• /
• v.16 no.7
• /
• pp.31-37
• /
• 1998

• ETRI Journal
• /
• v.43 no.6
• /
• pp.1004-1012
• /
• 2021

The collaboration productively interacting between multi-agents has become an emerging issue in real-world applications. In reinforcement learning, multi-agent environments present challenges beyond tractable issues in single-agent settings. This collaborative environment has the following highly complex attributes: sparse rewards for task completion, limited communications between each other, and only partial observations. In particular, adjustments in an agent's action policy result in a nonstationary environment from the other agent's perspective, which causes high variance in the learned policies and prevents the direct use of reinforcement learning approaches. Unexpected social loafing caused by high dispersion makes it difficult for all agents to succeed in collaborative tasks. Therefore, we address a paradox caused by the social loafing to significantly reduce total returns after a certain timestep of multi-agent reinforcement learning. We further demonstrate that the collaborative paradox in multi-agent environments can be avoided by our proposed effective early stop method leveraging a metric for social loafing.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 2006.07a
• /
• pp.282-282
• /
• 2006

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.493-498
• /
• 2008

We proposed "Table Mediator" which is a tabletop system for digital storytelling that uses web-retrieved information for the students' educational field trip. Students can perform their storytelling for their virtual pre-field trip to build up a sequential path as a story with web-retrieved documents, satellite images, geographical information, and group discussion. The proposed system was designed to lessen the limitation of individual interaction such as restricted viewpoint and biased inclination by group digital storytelling. Local interactions also have the limitation such as insufficient information and knowledge and the system supplied the rich live information such as subjective critiques or recently discovered history, or new updates for building a story that makes users arrange their own idea as a consistent story to lessen the limitation of the local interactions. The system can be used for various applications such as travel, education and other collaborative works with group interaction.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3266-3285
• /
• 2023

With the popularity of online learning, intelligent tutoring systems are starting to become mainstream for assisting online question practice. Surrounded by abundant learning resources, some students struggle to select the proper questions. Personalized question recommendation is crucial for supporting students in choosing the proper questions to improve their learning performance. However, traditional question recommendation methods (i.e., collaborative filtering (CF) and cognitive diagnosis model (CDM)) cannot meet students' needs well. The CDM-based question recommendation ignores students' requirements and similarities, resulting in inaccuracies in the recommendation. Even CF examines student similarities, it disregards their knowledge proficiency and struggles when generating questions of appropriate difficulty. To solve these issues, we first design an enhanced cognitive diagnosis process that integrates students' affection into traditional CDM by employing the non-compensatory bidimensional item response model (NCB-IRM) to enhance the representation of individual personality. Subsequently, we propose an affection-enhanced personalized question recommendation (AE-PQR) method for online learning. It introduces NCB-IRM to CF, considering both individual and common characteristics of students' responses to maintain rationality and accuracy for personalized question recommendation. Experimental results show that our proposed method improves the accuracy of diagnosed student cognition and the appropriateness of recommended questions.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.3
• /
• pp.643-651
• /
• 1996

In this paper, to provide the possibility of various multimedia services, especially the collaborative work using the centralized control feature of the Service Node, is studied. We focused on the telewriting as the upper layer application to confirm the basic functions of the collaborative work. For this, we implement and add telewriting collaborative work function to the Service Node emulator, and terminals operating in the Windows environment are also developed. Through the tests on the system, we verify the basic functions related to the collaborative work are performed adequately, and confirm the concept of Service Node can support various type of multimedia conference services.

• Asia pacific journal of information systems
• /
• v.8 no.3
• /
• pp.21-36
• /
• 1998

Group collaborative systems are recently emerging to support a group of users engaged in common tasks such as group decision making, engineering design, or collaborative writing. Simultaneously, as communications networks and distributed database systems become core underlying architecture of the organization, the need of collaborative systems are gaining more attentions from industry. In such collaborative systems, as the shared objects may evolve constantly or change for operational purposes, providing the users with synchronized and consistent views of the shared object and maintaining the consistency between shared object and replicated objects are important to improve the overall productivity. This paper provides an change management framework for the group collaborative systems to facilitate managing dependency relationships between shared objects and dependents, and coordinating change and propagation activities in distributed computing environments. Specifically, the framework adopts an object-oriented database paradigm and presents several object constructs capturing dependency management and change notification mechanisms. And the proposed framework accommodates both persistent dependents such as replicated data and transient dependents such as various user views in a single formalism. A prototype system is developed on a commercial object-oriented database management system called OBJECTSTORE using the C++ programming language.

• 한국HCI학회:학술대회논문집
• /
• 2006.02a
• /
• pp.443-448
• /
• 2006

본 논문에서는 고화질 협업 환경에서 다수의 고화질 영상들을 처리하기 위한 타일드 디스플레이(tiled display)기반의 확장성있는 디스플레이 구조를 제안한다. 제안하는 구조는 대형 고화질 디스플레이를 제어하기 위한 기술과 다수의 고화질 영상을 제한된 시스템 자원을 이용하여 효율적으로 디스플레이 하기위한 기술을 다룬다. 제안된 시스템은 영상의 획득/디코딩/디스플레이와 같은 가시화를 담당하는 Scalable Visualization Consumer 로 명명된 확장형 가시화 응용을 포함한다. 제안된 기법들을 토대로 구현된 확장형 가시화 시스템의 성능을 평가하고자 한다.

• Communications for Statistical Applications and Methods
• /
• v.11 no.1
• /
• pp.197-211
• /
• 2004

According to the increase of internet application, the marketing process as well as the research and survey, the education process, and administration of government are very depended on web bases. All kinds of goods and sales which are traded on the internet shopping malls are extremely increased. So, the necessity of automatically intelligent information system is shown, this system manages web site connected users for effective marketing. For the recommendation system which can offer a fit information from numerous web contents to user, we propose an automatic recommendation system which furnish necessary information to connected web user using statistical learning theory and bayesian fuzzy clustering. This system is called collaborative CRM in this paper. The performance of proposed system is compared with the other methods using real data of the existent shopping mall site. This paper shows that the predictive accuracy of the proposed system is improved by comparison with others.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.7
• /
• pp.768-774
• /
• 2016

This paper deals with a method to recommend the combination of items as a group according to similarity to handle application area such as fashion and cooking, while the previous methods recommend single item such as a book, music or movie. Collaborative filtering is a method to recommend an item selected by users with similar tendency based on similarity between users. In this paper, the proposed method generates a set of frequent items based on collaborative filtering and association rules and recommends a group by similarity between groups. To show the validity of the proposed method, experiments are performed with purchase data collected from e-commerce for four months.