• Journal of the Korean Society for information Management
• /
• v.30 no.4
• /
• pp.241-264
• /
• 2013

This study compared and analyzed weighted network centrality measures supported by Opsahl's tnet and Lee's WNET, which are free softwares for weighted network analysis. Three node centrality measures including weighted degree, weighted closeness, and weighted betweenness are supported by tnet, and four node centrality measures including nearest neighbor centrality, mean association, mean profile association, triangle betweenness centrality are supported by WNET. An experimental analysis carried out on artificial network data showed tnet's high sensitiveness on linear transformations of link weights, however, WNET's centrality measures were insensitive to linear transformations. Seven centrality measures from both tools, tnet and WNET, were calculated on six real network datasets. The results showed the characteristics of weighted network centrality measures of tnet and WNET, and the relationships between them were also discussed.

• Journal of the Korean Society for information Management
• /
• v.41 no.3
• /
• pp.511-533
• /
• 2024

This study aims to develop novel centrality measures applicable to networks that include both directional and weighted information, such as interlibrary loan networks and logistics transportation networks. While weighted PageRank has traditionally been used in such cases, experimental results reveal that it yields similar outcomes to neighborhood centrality, which measures local centrality. However, triangle betweenness centrality (TBC), despite assessing global centrality in weighted networks, does not consider link directions. To address these limitations, we propose two modified versions of the existing TBC measure: TBC-T for trust networks and TBC-F for flow networks. Applying these measures to two interlibrary loan networks, we find that TBC-T considers only the weights of inlinks, while TBC-F incorporates both inlink and outlink weights. These newly developed measures are expected to be useful for measuring node global centrality in weighted directed networks.

• The Journal of the Korea Contents Association
• /
• v.20 no.4
• /
• pp.658-672
• /
• 2020

The purpose of this study is to empirically compare statistical models for pre-service teachers' help networks. We identified similarities and differences based on the results of the binary and valued ERGM. Research questions are as follows: First, what are the similarities of factors influencing the binary/valued help network for pre-service teachers? Second, what are differences of factors influencing the binary/valued help network for pre-service teachers? We measured 42 pre-service teachers with focus on their help and friend networks, happiness, and personal characteristics. Results indicated that, first, the similar factors influencing the binary and valued help network of pre-service teachers were local dependencies (reciprocity, transitivity), similarity (major, gender), activity (early childhood education, negative emotion), popularity (early childhood education) and multiplicity (friend network). Second, the difference between factors affecting pre-service teacher's binary and valued help network was the effect of activity (physical education) and popularity (GPA, negative emotion). Based on these findings, we presented implications.

• Journal of the Korean Society for information Management
• /
• v.32 no.2
• /
• pp.7-23
• /
• 2015

While there are several measures for node centralities, such as betweenness and degree, few centrality measures for local centralities in weighted networks have been suggested. This study developed a generalized centrality measure for calculating local centralities in weighted networks. Neighbor centrality, which was suggested in this study, is the generalization of the degree centrality for binary networks and the nearest neighbor centrality for weighted networks with the parameter ${\alpha}$. The characteristics of suggested measure and the proper value of parameter ${\alpha}$ are investigated with 6 real network datasets and the results are reported.

• Journal of the Korean Society for information Management
• /
• v.25 no.4
• /
• pp.327-345
• /
• 2008

The main purposes of this study are to construct a Korean science network from journal contributions data of Korean researchers, and to analyze the structure and characteristics of the network. First of all, the association matrix of 140 scholarly domains are calculated based on the number of contributions in common journals, and then the Pathfinder network algorithm is applied to those matrix. The resulting network has several hubs such as 'Biology', 'Korean Language & Linguistics', 'Physics', etc. The entropy formula and several centrality measures for the weighted networks are adopted to identify the centralities and interdisciplinarity of each scholarly domain. In particular, the date hubs, which have several weak links, are successively distinguished by local and global triangle betweenness centrality measures.

• Proceedings of the Korean Society for Information Management Conference
• /
• 2013.08a
• /
• pp.27-30
• /
• 2013

공동연구 네트워크의 대표적인 사례인 공저 네트워크는 오랫동안 네트워크 분석의 대상으로 다루어져 왔다. 최근에는 가중 네트워크로서 공저 네트워크에 대한 연구가 활발해지면서 연구자의 영향력을 측정하려는 몇 가지 척도가 제안되었다. 이 연구에서는 공저 네트워크에서의 중심성을 측정하기 위해서 사용된 척도인 가중페이지랭크, 공동연구 h-지수와 공동연구 hs-지수, 복합연결정도중심성, c-지수에 대해서 비교 분석해본다.

• Journal of the Korean Society for information Management
• /
• v.31 no.3
• /
• pp.153-179
• /
• 2014

This study explores the characteristics of centrality measures for analyzing researchers' impact and structural positions in research collaboration networks. We investigate four binary network centrality measures (degree centrality, closeness centrality, betweenness centrality, and PageRank), and seven existing weighted network centrality measures (triangle betweenness centrality, mean association, weighted PageRank, collaboration h-index, collaboration hs-index, complex degree centrality, and c-index) for research collaboration networks. And we propose SSR, which is a new weighted centrality measure for collaboration networks. Using research collaboration data from three different research domains including architecture, library and information science, and marketing, the above twelve centrality measures are calculated and compared each other. Results indicate that the weighted network centrality measures are needed to consider collaboration strength as well as collaboration range in research collaboration networks. We also recommend that when considering both collaboration strength and range, it is appropriate to apply triangle betweenness centrality and SSR to investigate global centrality and local centrality in collaboration networks.

• Proceedings of the Korean Society for Information Management Conference
• /
• 2013.08a
• /
• pp.19-22
• /
• 2013

네트워크 분석이 확산되면서 여러 분야에서 다양한 중심성 척도가 개발되어 활용되고 있으나 가중 네트워크에서 지역중심성을 측정할 수 있는 척도로는 최근접이웃중심성 이외에는 거의 알려져 있지 않다. 최근접이웃중심성 척도는 동률값이 흔히 나타나므로 변별력이 낮다는 단점을 가지고 있다. 이 연구에서는 최근접이웃중심성 척도를 일반화한 이웃중심성 척도를 제안하고 가상 자료 및 실제 자료에 대해 적용하여 검증해보았다.

• KIPS Transactions on Computer and Communication Systems
• /
• v.6 no.7
• /
• pp.315-320
• /
• 2017

In this paper, I design a new scheme that is immune to malicious attack based on the weighted voting game. By using stochastic learning, the proposed scheme can revoke the certification of malicious node. Through the revocation process, the proposed scheme can effectively adapt the dynamic Mobile Ad hoc network situation. Simulation results clearly indicate that the developed scheme has better performance than other existing schemes under widely diverse network environments.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.6
• /
• pp.1284-1290
• /
• 2004

In market-basket analysis, weighted association rule(WAR) discovery can mine the rules which include more beneficial information by reflecting item importance for special products. However, when items are divided into more than one group and item importance for each group must be measured by different measurement or separately, we cannot directly apply traditional weighted association rule discovery. To solve this problem, we propose a novel methodology to discovery the weighted association rule in this paper In this methodology, the items should be first divided into sub-groups according to the properties of the items, and the item importance is defined or calculated only with the items enclosed to the sub-group. Our algorithm makes qualitative evaluation for network risk assessment possible by generating risk rule set for risk factor using network sorority data, and quantitative evaluation possible by calculating risk value using statistical factors such as weight applied in rule generation. And, It can be widely used for new model of more delicate analysis in market-basket database in which the data items are distinctly separated.