Michael J. Barber

Modularity and community detection in bipartite networks.

The modularity of a network quantifies the extent, relative to a null model network, to which vertices cluster into community groups. We define a null model appropriate for bipartite networks, and use it to define a bipartite modularity. The bipartite modularity is presented in terms of a modularity matrix B; some key properties of the eigenspectrum of B are identified and used to describe an algorithm for identifying modules in bipartite networks. The algorithm is based on the idea that the modules in the two parts of the network are dependent, with each part mutually being used to induce the vertices for the other part into the modules. We apply the algorithm to real-world network data, showing that the algorithm successfully identifies the modular structure of bipartite networks.

Detecting network communities by propagating labels under constraints

We investigate the recently proposed label-propagation algorithm (LPA) for identifying network communities. We reformulate the LPA as an equivalent optimization problem, giving an objective function whose maxima correspond to community solutions. By considering properties of the objective function, we identify conceptual and practical drawbacks of the label-propagation approach, most importantly the disparity between increasing the value of the objective function and improving the quality of communities found. To address the drawbacks, we modify the objective function in the optimization problem, producing a variety of algorithms that propagate labels subject to constraints; of particular interest is a variant that maximizes the modularity measure of community quality. Performance properties and implementation details of the proposed algorithms are discussed. Bipartite as well as unipartite networks are considered.

Spatial interaction modelling of cross‐region R&D collaborations: empirical evidence from the 5th EU framework programme*

The focus of this study is on cross-region R&D collaborations in Europe. We use data on collaborative R&D projects funded by the 5th EU Framework Programme (FP5). The objective is to identify separation effects - such as geographical or technological effects - on the constitution of cross-region collaborative R&D activities within a Poisson spatial interaction modelling framework. The results provide striking evidence that geographical factors are important determinants of cross-region collaboration intensities, but the effect of technological proximity is stronger. R&D collaborations occur most often between organizations that are located close to each other in technological space. Copyright (c) 2009 the author(s). Journal compilation (c) 2009 RSAI.

Distinct spatial characteristics of industrial and public research collaborations: evidence from the fifth EU Framework Programme

This study compares the spatial characteristics of industrial R&D networks to those of public research R&D networks (i.e. universities and research organisations). The objective is to measure the impact of geographical separation effects on the constitution of cross-region R&D collaborations for both types of collaboration. We use data on joint research projects funded by the fifth European Framework Programme (FP) to proxy cross-region collaborative activities. The study area is composed of 255 NUTS-2 regions that cover the EU-25 member states (excluding Malta and Cyprus) as well as Norway and Switzerland. We adopt spatial interaction models to analyse how the variation of cross-region industry and public research networks is affected by geography. The results of the spatial analysis provide evidence that geographical factors significantly affect patterns of industrial R&D collaboration, while in the public research sector effects of geography are much smaller. However, the results show that technological distance is the most important factor for both industry and public research cooperative activities.

R&D collaboration networks in the European framework programmes: data processing, network construction and selected results

We describe the construction of a large and novel data set on R&D collaboration networks in the first five EU Framework Programmes (FPs), examine key features and provide economic interpretations for our findings. The data set is based on publicly available raw data that pre-sents numerous challenges. We critically examine the different problems and detail how we have dealt with them. We describe how we construct networks from the processed data. The resulting networks display properties typical for large complex networks, including scale-free degree distributions and the small-world property. The former indicates the presence of net-work hubs, which we identify. Theoretical work shows the latter to be beneficial for knowl-edge creation and diffusion. Structural features are remarkably similar across FPs, indicating similar network formation mechanisms despite changes in governance rules. Several findings point towards the existence of a stable core of interlinked actors since the early FPs with inte-gration increasing over time. This core consists mainly of universities and research organisa-tions. The paper concludes with an agenda for future research.

Network of European Union-funded collaborative research and development projects.

We describe collaboration networks consisting of research projects funded by the European Union and the organizations involved in those projects. The networks are of substantial size and complexity, but are important to understand due to the significant impact they could have on research policies and national economies in the EU. In empirical determinations of the network properties, we observe characteristics similar to other collaboration networks, including scale-free degree distributions, small diameter, and high clustering. We present some plausible models for the formation and structure of networks with the observed properties.

Is the European R&D Network Homogeneous? Distinguishing Relevant Network Communities Using Graph Theoretic and Spatial Interaction Modelling Approaches

Barber M. J. and Scherngell T. Is the European R&D network homogeneous? Distinguishing relevant network communities using graph theoretic and spatial interaction modelling approaches, Regional Studies. This paper characterizes the geography of communities in the European research and development (R&D) network using data on R&D projects funded by the European Unions Fifth Framework Programme. Communities are sub-networks whose members are more tightly linked to one another than to other members of the network. The paper characterizes the communities by means of spatial interaction models, and estimates the impact of separation factors on the variation of cross-region collaboration activities in a given community at the level of 255 NUTS-2 regions. The results demonstrate that European R&D networks are not homogeneous, showing distinct, relevant substructures characterized by spatially heterogeneous community groups.

Searching for Communities in Bipartite Networks

Bipartite networks are a useful tool for representing and investigating interaction networks. We consider methods for identifying communities in bipartite networks. Intuitive notions of network community groups are made explicit using Newmans modularity measure. A specialized version of the modularity, adapted to be appropriate for bipartite networks, is presented; a corresponding algorithm is described for identifying community groups through maximizing this measure. The algorithm is applied to networks derived from the EU Framework Programs on Research and Technological Development. Community groups identified are compared using information‐theoretic methods.

Participations to European Framework Programs of higher education institutions and their association with organizational characteristics

This paper aims to analyze patterns of participation of higher education institutions (HEIs) to European Framework Programs (EU-FP) and their association with HEI characteristics, country and geographical effects. We have analyzed a sample of 2235 HEIs in 30 countries in Europe, derived from the European Tertiary Education Register (ETER), which has been matched with data on participations in EU-FPs in 2011 using the EUPRO database. Our findings identified (1) a high concentration of EU-FP participation in a small group of HEIs with high reputation; (2) the participation of non-doctorate awarding HEIs in EU-FPs is very limited despite the fact that they account for a significant share of tertiary student enrolments; (3) the number of participations tends to increase proportionally to organizational size, and is strongly influenced by international reputation; (5) there is limited evidence of significant country effects in EU-FP participations, as well as of the impact of distance from Brussels. We interpret these results as an outcome of the close association between HEI reputation and the network structure of EU-FP participants.

China’s landscape in oncology drug research: perspectives from research collaboration networks

OBJECTIVE Better understanding of Chinas landscape in oncology drug research is of great significance for discovering anti-cancer drugs in future. This article differs from previous studies by focusing on Chinese oncology drug research communities in co-publication networks at the institutional level. Moreover, this research aims to explore structures and behaviors of relevant research units by thematic community analysis and to address policy recommendations. METHODS This research used social network analysis to define an institutions network and to identify a community network which is characterized by thematic content. RESULTS A total of 675 sample articles from 2008 through 2012 were retrieved from the Science Citation Index Expanded (SCIE) database of Web of Science, and top institutions and institutional pairs are highlighted for further discussion. Meanwhile, this study revealed that institutions based in the Chinese mainland are located in a relatively central position, Taiwans institutions are closely assembled on the side, and Hong Kongs units located in the middle of the Chinese mainlands and Taiwans. Spatial division and institutional hierarchy are still critical barriers to research collaboration in the field of anti-cancer drugs in China. In addition, the communities focusing on hot research areas show the higher nodal degree, whereas communities giving more attention to rare research subjects are relatively marginalized to the periphery of network. CONCLUSIONS This paper offers policy recommendations to accelerate cross-regional cooperation, such as through developing information technology and increasing investment. The brokers should focus more on outreach to other institutions. Finally, participation in topics of common interest is conducive to improved efficiency in research and development (R&D) resource allocation.