Earticle

본 연구에서는 협력 네트워크 차원에서의 북한이탈주민의 정착 지원 프로그램 구조를 살펴보기 위한 연구이다. 북한이탈주민의 누적 인구분포가 지속적으로 증가함에 따라 이를 수용할 수 있는 시설과 프로그램의 필요성이 제기되고 있다. 각 지방정부들은 지역 설정에 맞는 서비스를 제공하기 위해 지역별 사회복지단체와 연대하여 북한이탈주민들의 서비스를 제공하고자 노력하고 있다. 따라서 본 연구에서는 각 협력기관들간의 북한이탈주민들의 정착지원과 관련된 정보교환망의 협력 네트워크 구조를 살펴봄과 동시에 이들의 연계패턴을 분석하고자 한다. 이를 측정하기 위해 네트워크 속성, 네트워크 결속, 중앙성(centrality), 결집집단(cohesion), 구조적 등위성(CONCOR) 등을 살펴보았다. 연구 결과 향후 정보교류 및 제공의 차원에서 보다 낳은 서비스 제공을 위해서는 이들의 관계를 재조정할 필요가 있고, 이에 이를 조정할 수 있는 책임있는 기관 및 중재자가 요청됨을 발견하였다. 따라서, 중앙정부, 지방정부, 민간단체, 지역기업 및 사회적 기업 간의 협력 네트워크에 있어 사회통합 모니터링 센터를 중심으로 소통영역, 신뢰영역, 협력영역, 기회 영역별로 정보교환을 원활하게 할 수 있게 구축되어야 한다. 향후에는 정부주도 민간 보조형으로 사회 정착 지원에 상당한 역할을 보완할 필요가 있음을 확인하였다. 따라서 이러한 밀도, 중앙성, 응집성, 구조적 등위성 등을 분석함으로써 북한이탈주민의 정착지원의 네트워크의 중심과 협력성에 대한 정책적 시사점을 도출 및 제안하였다.

This study seeks to illustrate the structure of settlement support program for North Korean defectors applying the concept of collaborative network. Network features, unity, centrality, cohesion and structural equivalence are measured which would explain collaboration network structure on information exchange among settlement support programs for North Korean defectors and collaborating organizations as to identify the structural characteristics. It is found that reliable organization or mediator is in need for coordination among themselves. Therefore, collaborative network among central government, local government and social corporation should be established around social cohesion monitoring center with an emphasis on information exchange through communication, trust, collaboration and opportunity. Also, it turns out that settlement support system should be improved in order to seek for government taking the initiative and private sector providing assistance. By analyzing density, centrality, cohesion and structural equivalence, policy implications are drawn for the core and collaboration of settlement support network.

Recent advancements in Deep Learning (DL) techniques for fire detection have mitigated various ecological, economic, and environmental impacts. However, single models in existing literature often perform poorly due to their inability to capture relevant features with limited data. In this study, we therefore develop an innovative framework for effective fire detection using teacher-student collaborative knowledge distillation. The framework comprises two main components: the teacher model, which leverages a pretrained InceptionV3, and a customized student model designed to inherit the knowledge from the pretrained InceptionV3 to a pruned student model. The proposed network is evaluated and compared against several competitive techniques using two datasets, MAFire and Yar, with different evaluation metrics, offering higher performance and lower computational cost.

This study proposes novel recommender systems which collect qualitative and emotional information for recommendation engine in order to mitigate the limitations of typical collaborative filtering (CF) algorithm. The proposed model selects subgroups of users in Internet community through social network analysis (SNA), and then performs clustering analysis using the information about subgroups. Finally, it makes recommendations using cluster-indexing CF based on the clustering results. This study tries to use the cores in subgroups as an initial seed for a conventional clustering algorithm. This model chooses five cores which have the highest value of degree centrality from SNA, and then performs clustering analysis by using the cores as initial centroids (cluster centers). Then, the model amplify the impact of friends in social network in the process of cluster-indexing CF. The experimental results show the proposed model outperforms conventional CF and other comparative models with statistical significance.

재난안전 협력연결망이 기관간에 지속적으로 형성되고 유지되기 위해서는 네트워크 존속에 대한 연구가 수반될 필요가 있다. 즉, 과연 그 연결망이 어떻게 형성된 것일까? 어떤 이유로 형성되고 지속되는 결과를 가져올 수 있을까? 등에 대해 설명할 방법은 사실상 없었으며, 이는 매우 어려운 일이었다. 이러한 맥락에서 본 연구는 한국재난안전 협력연결망의 존속을 분석하고 예측할 수 있는 ERGM방법에 기초하여 분석을 시도하였다. 분석결과 우선, 호혜성 변수가 통계적으로 유의미하였고 추정치 역시 높게 나타났다는 점이다. 즉, 호혜적 관계하에서 재난안전 연결망은 형성 및 지속성을 띨 수 있다는 것이다. 이는 재난안전 연결망을 지속하기 위해 호혜적인 관계에 대한 적실한 보상과 여건의 확보가 필요하다. 둘째, 확산경향은 연구가설인 정(+)의 관계와 대립되는 부(-)의 관계성으로 확인되었다. 특히, 선행연구들에서 확산경향은 변수의 방향이 본 연구와 동일한 (-)의 방향으로 확인되었다. 비록 이 변수가 이론가설과는 다를지라도 장기적으로 신뢰의 지속성에 대한 상실의 관점에서 네트워크의 지속성에 부의 영향을 미치는 경쟁가설이 제기될 수도 있다. 따라서, 확산경향에 대한 검증은 향후 많은 연구를 통해서 검증하고 타당성을 논의해야할 필요가 있다. 셋째, 인기도는 역시 모든 모형에서 통계적 유의성을 확인할 수 있었고, 모든 모형에서 가설방향과 일치하는 결과를 나타내었다. 향후 정부 및 지자체는 재난관련 기관들 중에서 사회적 명망이 높은 기관들을 선별하고 이들의 적극적인 재난안전 연결망 참여를 촉구할 필요가 있다. 마지막으로, 모든 모형에서 확인되지는 않았으나 이행성 역시 가설의 방향과 일치되는 결과를 확인해볼 수 있었다. 즉, 이행성이 존재하는 경우에 한국의 재난안전 협력연결망은 형성될 가능성이 높으며, 또한 이행성이 증대한다면 재난안전 협력연결망은 지속될 가능성이 높아지게 된다는 것이다.

This paper examines to the continuity of the network which is necessary for continual formation and maintenance of Disaster Management Collaborative Networks between agencies. It is difficult to explain the formation and continuity of networks, so that this paper tried to analyse Korean Disaster Management Network based on the ERGM method in order to examine existence of the network. Findings of this research are following. First, the result shows that the reciprocity variable is significantly significant, which means that disaster management networks are formed and continued in reciprocity relations. Second, is also statistically significant in all models, and the result shows the popularity accords with hypotheses. Therefore, governments need to sort out the reputed agencies, and induce agencies to participate in networks. Third, it is found that transitivity also accords with hypothesis, although transitivity could not be confirmed in all models. However, the diffusion tendency is confirmed that it has negative correlation with the hypothesis, along with the results of preceding researches.

이제까지 개발된 많은 어린이 대상의 독서교육 프로그램은 사서모임이나 교사모임, 어린이책 시민단체, 연구자 집단, 독서관련 사기업, 출판업계 등 직무와 관련된 단체별로 진행되어 왔다. 따라서 개별 단체 내에서는 독서교육에 대한 성과와 노하우가 상당히 축적되고 공유되었으나, 단체 간의 상호교차적인 정보와 기술의 교환은 활발하지 못했다. 그 결과 다른 전문가 집단의 성과를 어린이 사서가 공유하고, 자신의 경험에 대해 다른 전문가 집단과 논의할 수 있는 기회가 희박했다. “책수리마수리"는 책을 매개로 어린이와 청소년을 만나는 다양한 직종의 사람들이 모여 만드는 독서교육 프로그램이다. “책수리마수리" 프로젝트가 시범적으로 이루어진 2010년의 경우, 연구자, 공공도서관 사서, 학교도서관 사서, 초등학교 교사, 그림책 작가, 코디네이터 사서가 한 팀이 되어 ‘크기'를 주제로 다양한 독서교육 프로그램을 만들었다. 전문가 협력을 통해 어린이 사서는 잘 만들어진 프로그램을 이식하는 전달자를 넘어서 프로그램의 개발에서부터 진행까지 주체적으로 관여하며, 다양한 분야의 전문성을 가진 인적자원의 네트워크와 함께 활동할 수 있게 된다.

Bookmagic is a new reading program that encourages children of 5-15 ages to enjoy reading. Bookmagic provides free downloadable resources such as pre-reading and after-reading activities, booklists, posters and awards. Since a number of reading programs have already been developed, published and used, Bookmagic is not a toally new project. However, it is distinguished by the process of creating the program. While most of the reading programs available in Korea were developed by individual occupations(academic experts, private enterprises, civic organisations, associations of librarians or teachers), Bookmagic was designed by a team of various professionals including a picture book author, an academic researcher, a primary school teacher, a school librarian and two public librarians. Working in a partnership with other professionals, participant librarians had a unique opportunity to develop expertises on reading education as a creator of a program rather than as a deliverer.

본 연구에서는 국내에서 재난관리에 있어 지방정부와 민간과의 네트워크를 지역특성별로 구성하는 방안에 대해 제시하였다. 현재 우리들이 살고 있는 과학기술이 발달한 현대 산업사회를 ‘위험사회’ 라는 개념으로 정의하기도 한다. 최근 재난의 유형이 다변화되고 피해규모가 증가됨에 따라 본래의 정부주도하의 재난대응보 다는 협업과 협력을 통한 재난관리가 필요하다. 국외에는 이러한 재난대응을 위한 협력구축네트워크를 광 역방재거점 등을 설치·운영함으로써 실질적이고 효율적인 협력을 구축하고 운영 중이다. 하지만, 국내에서 는 협업 및 협력에 대한 매뉴얼 등을 통한 소프트적인 체계구축에 중점을 두고 대응을 하고 있다. 본 연구 에서는 국외사례를 통해 실질적인 거점형태의 민관네트워크를 구축하고, 해당지역에 적합한 운영방안을 선 택하는 데에 활용 할 수 있을 것이다.

이 연구는 사회네트워크분석(SNA)을 사용하여 지역 내 다문화가족지원기관 간 협력네트워크의 구조를 파악하는데목적을 두고 있다. 이를 위해 부산지역의 다문화가족지원센터, 공공기관, 공공도서관, 사회복지관, 기타 민간 영역의다문화 관련 기관을 대상으로 정보 및 자료교환, 인적·물적 교류, 재정지원, 공동사업, 행사참여에 대한 관계데이터를수집하고, 이를 토대로 협력 네트워크의 구조적 특성을 분석하였다. 분석결과, 협력네트워크는 전반적으로 활성화되어있지 않았으며, 일부 협력 유형의 경우 기능이나 성격, 역할이 유사한 기관들끼리만 협력이 이루어지고 있어 상당히제한적인 네트워크 구조를 보이고 있었다. 또한 다문화 관련 기관들 중 공공기관과 다문화가족지원센터가 주로 네트워크에서 중심적인 역할을 수행하며, 이들 기관은 다른 기관들과 상대적으로 골고루 연결되어 있는 반면, 도서관의 경우도서관 간 협력만 이루어지거나, 협력활동이 아예 없는 경우가 많았다. 따라서 도서관을 통한 효과적인 다문화 서비스를위해 다른 기관들과의 협력과 교류를 활성화할 필요가 있다.

This study attempts to analyze the structure of collaborative network among multicultural agencies in Busan area. The survey was conducted with such agencies as multicultural family support centers, regional government offices, public libraries, social welfare centers, and other NGO-supported agencies, in order to analyze the structural features of the network. The results of the study indicate that cooperative networks are hardly in action except for the one among agencies which share their function, roles, and characteristics in common, that regional government offices and multicultural family support centers have high density networks with other agencies and performed major roles. Public libraries, however, only collaborate with other libraries, or did not participated in the networks, which indicates that endeavors of public libraries to cooperate with other multicultural agencies is needed.

본 연구의 목적은 협력 학습에서 소셜 네트워크 서비스를 활용했을 때, 구성원들의 협력 능력, 협력만족도, 그리고 집단내 상호작용에 어떤 영향을 미치는지를 분석하는데 있었다. 협력 과정에서 소셜 네트워크 서비스 중의 하나인 카카오톡을 활용하여 과제를 수행한 집단은 그렇지 않은 집단에 비해 협력 능력과 협력 만족도의 정도가 유 의미하게 높았다(p<.05). 카카오톡을 활용한 집단을 대상으로 생성한 담화의 양과 내용을 분석한 결과 담화의 양은 협력 능력이나 협력 만족도에 영향을 미치지 않는 것으로 나타났다.

The purpose of this study was to analyze the effects of social network service on the collaborative skills, collaborative satisfaction, and interaction within groups in collaborative learning. The group that used KakaoTalk, one of social network service for working on the collaborative task in the course exhibited higher collaborative skills and collaborative satisfaction (p<.05) than the group that did not use KakaoTalk. When analyzing the amount and the content of the messages produced by the group that used KakaoTalk, the amount of messages did not have an impact on the collaborative skills and collaborative satisfaction.

This paper presents u-COEX, Ubiquitous-Collaborative Online shopping EXecution system, for small- and mediam- sized business enterprises, based on EPCglobal network. The system is taking advantage of RFID technology promises to optimize the critical processes in the Supply Chain Management. The system consists of five major functions: integrated order management, realtime monitoring and analysis system of sales and inventory, decision support system, integrating with EPCglobal and RFID technology, and u-catalog feature. The prototype implementation was developed for mass electronic market complex and the result revealed the feasibility to be applicable to real market.

Collaborative Digital Reference Service(CDRS) has spread worldwide as a tool of obtaining high-quality information in knowledge information society. KISTI formed CDRS collaboration model called Question포인트+ with 4 institutions in Korea. In Question포인트+ which is also linked to the Global Network, the domestic and oversea information specialists answer a variety of questions in response of the requests of users on the web. The implementation of Question포인트+ is significant as the first domestic CDRS collaboration model, and contribute to management, dissemination, and activation of CDRS in Korea.

This study investigates the social network among authors to improve the quality of Panel researches. Korea Health Panel (KHP), implemented by the collaborative work between KIHASA (Korea Institute for Health and Social Affairs) and NHIC (National Health Insurance Service) since 2008, provides a critical infrastructure for policy making and management for insurance system and healthcare service. Using bibliographic data extracted from academic databases, eighty articles were extracted in domestic and international journals from 2008 to 2014, April. Data were analyzed by NetMiner 4.0, social network analysis software, to identify the extent to which authors are involved in healthcare use research and the patterns of collaboration between them. Analysis reveals that most authors publish a very small number of articles and collaborate within tightly knit circles. Centrality measures confirm these findings by revealing that only a small percentage of the authors are structurally dominant, and influence the flow of communication among others. It leads to the discovery of dependencies between the elements of the co-author network such as affiliates in health panel communities. Based on these findings, we recommend that Korea Health Panel could benefit from cultivating a wider base of influential authors and promoting broader collaborations.

Companies have made an investment of cost and time to optimize processing of a new business model in a cloud environment, applying collaboration technology utilizing business processes in a social network. The collaborative processing method changed from traditional BPM to the cloud and a mobile cloud environment. We proposed a collaborative system for operating processes in social networks using MOXMDR-DAI+ (eXtended Metadata Registry-Data Access & Integration based multimedia ontology). The system operating cloud-based collaborative processes in application of MOXMDR-DAI+, which was suitable for data interoperation. MOXMDR-DAI+ applied to this system was an agent effectively supporting access and integration between multimedia content metadata schema and instance, which were necessary for data interoperation, of individual local system in the cloud environment, operating collaborative processes in the social network. In operating the social network-based collaborative processes, there occurred heterogeneousness such as schema structure and semantic collision due to queries in the processes and unit conversion between instances. It aimed to solve the occurrence of heterogeneousness in the process of metadata mapping using MOXMDR-DAI+ in the system. The system proposed in this study can visualize business processes. And it makes it easier to operate the collaboration process through mobile support. Realtime status monitoring of the operation process is possible through the dashboard, and it is possible to perform a collaborative process through expert search using a community in a social network environment.

The goal of deep learning is to extract complex features from multidimensional data use the features to create models that connect input and output. Deep learning is a process of learning nonlinear features and functions from complex data, and the user data that is employed to train deep learning models has become the focus of privacy concerns. Companies that collect user’s sensitive personal information, such as users’ images and voices, own this data for indefinite period of times. Users cannot delete their personal information, and they cannot limit the purposes for which the data is used. The study has designed a deep learning method that employs privacy protection technology that uses distributed collaborative learning so that multiple participants can use neural network models collaboratively without sharing the input datasets. To prevent direct leaks of personal information, participants are not shown the training datasets during the model training process, unlike traditional deep learning so that the personal information in the data can be protected. The study used a method that can selectively share subsets via an optimization algorithm that is based on modified distributed stochastic gradient descent, and the result showed that it was possible to learn with improved learning accuracy while protecting personal information.

By a variety of accounts, the number and scope of interorganizational collaborations have grown rapidly in many industries, especially in the large assembly industries such as shipbuilding, automobile, machine. Despite the emphasis on developing cooperative subcontracting relationships, many relationships between powerful contractor and upstream subcontractors are still adversarial. Many scholars argues that developing relationship commitment can have positive consequences for the participants in the interfirm relationships. But in the world of practice, the application of relationship commitment to the substantial subcontracting networks have many limitations.This study proposed knowledge－sharing routine as mediator for interfirm relationships in addition to relationship commitment. Empirical results shows that knowledge－sharing routine as mediator has increased relationship performance in the subcontracting networks. Therefore, the introduction of formal knowledge－sharing routine in the subcontracting networks has important role in the construction of collaborative network for creation of key knowldege asset.

This paper focuses on the relationship between the characteristics of network and the productivity of scientists, which is rarely examined in previous studies. Utilizing a unique dataset from the Korean Citation Index (KCI), we examine the overall characteristics of the research network (e.g. distribution of nodes, density and mean distance), and analyze whether the network centrality is related to the scientific productivity. According to the results, firstly we have found that the collaborative research network of the Korean academics in the field of statistics and computer science is a scale-free network. Secondly, these research networks show a disciplinary difference. The network of statisticians is denser than that of computer scientists. In addition, computer scientists are located in a fragmented network compared to statisticians. Thirdly, with regard to the relationship between the researchers' network position and scientific productivity, a significant relation and their disciplinary difference have been observed. In particular, the degree centrality is the strongest predictor for the scientists' productivity. Based on these findings, some policy implications are put forward.

Concurrent Engineering (CE) is a popular method employed in product development. It treats the whole product design process by the consideration of product quality, cost, rate of progress, and demands of customers. The development of computer and network technologies provides a strong support to the realization of CE in practice. Aiming at the characteristics of CE and network collaborative design, this paper built network collaborative design system frame. Through the analysis of the network collaborative design modes based on CE, this paper provided a novel network collaborative design integration model. This model can integrate the product design information, design process, and knowledge. Intelligent collaboration was considered in the proposed model. The study showed that the proposed model considered main factors such as information, knowledge, and design process in collaborative design. It has potential application in CE fields.

Measuring performance of public goals can be highly complicated considering that, in the achievement of those goals, governments are no longer directly involved in the delivery of goods and services. Hybridization of public functions, purposes and processes has created new challenges by fragmenting public bureaucratic functions into the hands of multiple players, thus complicating measuring and reporting performance outcomes. This article presents the results of a study that examined performance indicators across multiple public agencies in a large municipal county. The purpose was to assess whether performance measures link organizational collective goals. It specifically addresses the following question: How does the governance of multi-actor networks influence the design of performance measures? The study emphasizes a need for the design of interlinked performance measures in a large networked environment. The findings indicate that design of performance measures remains insular to the organization and does not reflect the hybridized collaboration between various entities.

Many researchers claim that the interpersonal aspect of peer familiarity is important and has a crucial role in performing collaborative tasks. However, this factor has not been studied thoroughly until now. This paper seeks to analyze the relationship between the linguistic aspect (writing performance and written interaction) and the interpersonal aspect (writers' familiarity among group members) in producing better writing performance on weblogs using Social Network Analysis (SNA). SNA is able to examine intangible aspects, like network or relationships among people, and it has been adopted by a wide variety of researchers. SNA shows relationships among people and calculates a writer' written interactions and familiarity among group members in blogging that enables collaborative online weblog writing. This study asked and answered three key questions: how peer familiarity relates to written interactions in weblogs, what are the correlations between peer familiarity, written interactions and writing performance, and how students' perspectives and attitudes toward peer familiarity differ in weblogs. The results of this research suggest that the interpersonal factor of peer familiarity plays an important role in fostering written interactions and improving writing ability.