• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.141-148
• /
• 2022

Recently, reinforcement learning combined with deep neural network technology has achieved remarkable success in various fields such as board games such as Go and chess, computer games such as Atari and StartCraft, and robot object manipulation tasks. However, such deep reinforcement learning describes states, actions, and policies in vector representation. Therefore, the existing deep reinforcement learning has some limitations in generality and interpretability of the learned policy, and it is difficult to effectively incorporate domain knowledge into policy learning. On the other hand, dNL-RRL, a new relational reinforcement learning framework proposed to solve these problems, uses a kind of vector representation for sensor input data and lower-level motion control as in the existing deep reinforcement learning. However, for states, actions, and learned policies, It uses a relational representation with logic predicates and rules. In this paper, we present dNL-RRL-based policy learning for transportation mobile robots in a manufacturing environment. In particular, this study proposes a effective method to utilize the prior domain knowledge of human experts to improve the efficiency of relational reinforcement learning. Through various experiments, we demonstrate the performance improvement of the relational reinforcement learning by using domain knowledge as proposed in this paper.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.5
• /
• pp.60-72
• /
• 1996

In this paper, a enw model-base three-dimensional (3-D) object recognition mehtod using hopfield network is proposed. To minimize deformation of feature values on 3-D rotation, we select 3-D shape features and 3-D relational features which have rotational invariant characteristics. Then these feature values are normalized to have scale invariant characteristics, also. The input features are matched with model features by optimization process of hopjfield network in the form of two dimensional arrayed neurons. Experimental results on object classification and object matching with the 3-D rotated, scale changed, an dpartial oculued objects show good performance of proposed method.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.11b
• /
• pp.363-367
• /
• 2002

일반 중소기업들의 업무는 수작업으로 데이터를 유지하고 있는 수준에 머물고 있는 실정이다. 인건비를 포함한 제반비용의 절감과 업무능률의 향상이 필요하다. 데이터 관리 및 유지를 위해서 효율적이고 신속 정확한 작업처리를 위해 데이터 베이스를 이용한 통합 자원관리 시스템이 구축이 필요하다. 이 프로젝트는 업무 효율을 높이기 위하여 데이터 베이스를 이용한 통합 관리 시스템을 개발하고 용이한 자료 추출을 위한 자료추출 엔진 및 출력 폼 개발 및 Network구성을 통한 분산환경에서의 사무자동화 구현을 목표로 개발하였다.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.41 no.11
• /
• pp.39-45
• /
• 2004

This paper proposes an algorithm of address analysis for the next-generation networks. That is, we present integrated address analysis of the Numbering Plans of public telecommunication networks such as those of the ATM network. The purpose of implementing software for integrated address analysis is to actively and flexibly cope with a large change of information while operating the system and the network. By designing and implementing a database structure and an address analysis algorithm, interworking is possible among part (local) networks that each have a different Addressing Plan within an integrated network. Proposed relational model database structure has effect that do to manage address information of complicated network efficiently. Also, apply in real environment and proved excellency through interworking test to verify mechanism that propose.

• Korean Institute of Interior Design Journal
• /
• v.22 no.6
• /
• pp.79-87
• /
• 2013

The purpose of this study clarifies an expression characteristic of the phenomenal space. The architecture is an interface between human and nature. Nature presents herself as phenomena. Thus, the phenomenal space should be approached as the essence of architecture that is to accommodate nature. Phenomenon is related to everyday life and shares flow naturally within it. The phenomenon and everyday life form a relationship through the mediating elements that are time, place, and image. If these mediating elements are developed as spatialized elements, time becomes the converse, place becomes the overlap, and shape becomes the revealing. Also, spatial components that are substituted with these elements are void/solid, form, and materials. The relational characteristics of phenomenal space can be identified through these, and such characteristics are one-ness, continuity, and coincidence of opposites. Phenomenal space is expressed with spatial tones and accepted as spatial atmospheres. For the analysis, 15 works of swiss architects were selected to which spatial elements were applied. And It were composed that analysis by arranging these components as the relational network found that expression characteristics. Trough the analysis, It was found that expression characteristics of phenomenal space of the architecture of Swiss architects were prototypicality, primitiveness, and originality. As a results, It is considered that the role of the space that contains the value of everyday life, the value of the phenomenon is necessary.

• KIPS Transactions on Computer and Communication Systems
• /
• v.5 no.5
• /
• pp.117-126
• /
• 2016

Recently, increasing utilization of Big Data or Social Network Service involves the increases in demand for NoSQL Database that overcomes the limitations of existing relational database. A forensic examination of Relational Database has steadily researched in terms of Digital Forensics. In contrast, the forensic examination of NoSQL Database is rarely studied. In this paper, We introduce Redis (which is) based on Key-Value Store NoSQL Database, and research the collection and analysis of forensic artifacts then propose recovery method of deleted data. Also we developed a recovery tool, it will be verified our recovery algorithm.

• Journal of Convergence for Information Technology
• /
• v.10 no.7
• /
• pp.41-48
• /
• 2020

Recently, the spreading business of AMI (Advanced Metering Infrastructure) remote metering systems in various regions of the country has been activated, and it provides various metering functions such as two-way communication and security plan functions for power demand management. Current AMI system is difficult to analyze based on the existing RDB(Relational Database) due to the increase in the size of new internal IoT devices and networks. This study proposes a new GDB(Graph Database) based failure analysis method that utilizes existing RDB data. It analyzes the correlation of new failure patterns through accumulated data such as internal thresholds and status values. As a result of GDB-based simulation, it was confirmed that RDB can predict to a new obstacle pattern that was difficult to analyze.

• Journal of Contemporary Eastern Asia
• /
• v.19 no.1
• /
• pp.8-30
• /
• 2020

Most literature on the "comfort women" social movement focuses on the case of Korea. These works tend to transpose the meanings generated by South Korean organizations onto the transnational network, assuming certain homogeneity of repertoires and identities among the different social actors that comprise this network. Even though there is some degree of consensus about demands, repertoires, and advocacy strategies at the international level, does this same uniformity exist at the national level? In each country, what similarities and differences are present in the laboratories of ideas, relationships, and identities of social actors in the network? Symbolically and politically, do they challenge their respective societies in the same way? This article compares this social movement in South Korea, China, and Taiwan. My main argument is that the constitutive base for this transnational network is the domestic actions of these organizations. It is in the domestic sphere that these social actors reinforce their agendas, reinvent their repertoires, transform their identities, and expand their submerged networks, allowing national movements to retain their latency and autonomy. Following Melucci's relational approach to the study of social movements, this research is based on a qualitative analysis of institutional documents, participant observation, and open-ended interviews with members of the main social actors.

• Journal of Biomedical Engineering Research
• /
• v.18 no.1
• /
• pp.17-24
• /
• 1997

We have developed a patient monitoring system including module-based bedside monitors, interbed network, central stations, clinical workstations, and DB servers. A bedside monitor with a color LCD can accommodate up to 3 module cases and 21 different modules. Six different physiological parameters of ECG, respiration, invasive blood pressure, noninvasive blood pressure, body temperature, and arterial pulse oximetry with plethysmoyaph are provided as parameter modules. In a single bedside monitor, modules and a module controller communicate with IMbps data rate through an intrabed network based on RS-485 and HDU protocol. At the same time, it communicates with other bedside monitors and central stations through interbed network based on 1 OMbps Ethernet and TCP/IP protocol. Central stations using 20" color CRT monitors can be connected with many bedside monitors and they display 18 channels of waveforms simultaneously. Clinical workstations are used mainly for the review of patient datE In order to accommodate more advanced data management capabilities such as 24-hour full disclosure, we have developed a relational database server dedicated to the patient monitoring system. Software for bedside monitor, central station, and clinical workstation fully utilizes graphical user interface techniques and all functions are controlled by a rotate/push button on the bedside monitor arid a mouse on the central station and clinical workstation. The entire system satisfies the requirements of AAMI and ANSI standards in terms of electrical safety and performances.nces.

• The Journal of Information Systems
• /
• v.25 no.1
• /
• pp.201-223
• /
• 2016

Purpose As social network games (SNG) enjoy rapid growth in the market and become a major sector of the gaming industry, it is of great interest to examine the how users continuously use SNG. In SNG, the users' social interaction is the most prominent advantage of the social network, as well as the entertainment afforded by the game. This study explores the relationship between altruism, which is considered the most prominent characteristic of SNS, and the continuance usage intention, as well as the moderating role of social capital. Based on social capital theory and organizational citizenship behavior, this research model considers social bonding and bridging that are divided by social capital. Design/methodology/approach An AMOS analysis based on survey data from 223 SNG users indicated that SNG with greater altruism enhance social capital (social bonding, social bridging), which is related to the user's satisfaction and the continuance intention of SNG. Findings Social bonding is positively related to the user's satisfaction with SNG. In other words, social bridging positively affects the continuous usage intention of SNG. These findings help managers in developing and implementing altruistic relationships and social capital for continuous usage of SNG.