Results 1-10 of 10.
((uid:50022055))

Bookmark and Share    
Full Text
Peer Reviewed
See detailThe Latent Topic Block Model for the Co-Clustering of Textual Interaction Data
Berge, Laurent UL; Bouveyron, Charles; Corneli, Marco et al

in Computational Statistics and Data Analysis (2019), 137

Detailed reference viewed: 64 (3 UL)
Full Text
Peer Reviewed
See detailUnintended triadic closure in social networks: The strategic formation of research collaborations between French inventors.
Carayol, Nicolas; Berge, Laurent UL; Cassi, Lorenzo et al

in Journal of Economic Behavior and Organization (2019), 163

Detailed reference viewed: 49 (3 UL)
Full Text
Peer Reviewed
See detailHow do inventor networks affect urban invention?
Berge, Laurent UL; Carayol, Nicolas; Roux, Pascale

in Regional Science & Urban Economics (2018), 71

Social networks are expected to matter for invention in cities, but empirical evidence is still puzzling. In this paper, we provide new results on urban patenting covering more than twenty years of ... [more ▼]

Social networks are expected to matter for invention in cities, but empirical evidence is still puzzling. In this paper, we provide new results on urban patenting covering more than twenty years of European patents invented by nearly one hundred thousand inventors located in France. Elaborating on the recent economic literatures on peer effects and on games in social networks, we assume that the productivity of an inventor’s efforts is positively affected by the efforts of his or her partners and negatively by the number of these partners’ connections. In this framework, inventors’ equilibrium outcomes are proportional to the square of their network centrality, which encompasses, as special cases, several well-known forms of centrality (Degree, Katz-Bonacich, Page-Rank). Our empirical results show that urban inventors benefit from their collaboration network. Their production increases when they collaborate with more central agents and when they have more collaborations. Our estimations suggest that inventors’ productivity grows sublinearly with the efforts of direct partners, and that they incur no negative externality from them having many partners. Overall, we estimate that a one standard deviation increase in local inventors’ centrality raises future urban patenting by 13%. [less ▲]

Detailed reference viewed: 102 (10 UL)
Full Text
Peer Reviewed
See detailNetwork proximity in the geography of research collaboration
Berge, Laurent UL

in Papers in Regional Science (2017), 96(4),

This paper deals with the questions of how network proximity influences the structure of inter-regional collaborations and how it interacts with geography. I first introduce a new, theoretically grounded ... [more ▼]

This paper deals with the questions of how network proximity influences the structure of inter-regional collaborations and how it interacts with geography. I first introduce a new, theoretically grounded measure of inter-regional network proximity. Then, I use data on European scientific co-publications in the field of chemistry between 2001 and 2005 to assess those questions. The main findings reveal that inter-regional network proximity is important in determining future collaborations but its effect is mediated by geography. Most importantly, a clear substitution pattern is revealed showing that network proximity mainly benefits international collaborations. [less ▲]

Detailed reference viewed: 140 (15 UL)
Full Text
Peer Reviewed
See detailSoftware patents and scientific publications
Berge, Laurent UL; Doherr, Thorsten; Hussinger, Katrin UL

in Academy of Management Proceedings, 2017 (2017)

Detailed reference viewed: 74 (5 UL)
Full Text
Peer Reviewed
See detailCentrality of regions in R&D networks: A new measurement approach using the concept of bridging paths
Berge, Laurent UL; Wanzenbock, Iris; Scherngell, Thomas

in Regional Studies (2017), 51(8), 1165--1178

This paper aims at introducing a novel measure of regional centrality in the context of R&D networks. We first demonstrate some substantial problems of SNA-based centrality measures to cope with regional ... [more ▼]

This paper aims at introducing a novel measure of regional centrality in the context of R&D networks. We first demonstrate some substantial problems of SNA-based centrality measures to cope with regional R&D networks in a meaningful way. Then, we introduce a new measurement approach of regional network centrality based on the concept of inter-regional bridging paths (indirect connections at the regional level). We show that the formal definition of our regional bridging centrality measure can be expressed in terms of three simple components: the participation intensity of a region in inter-regional R&D collaborations, the relative outward orientation in terms of all established links and the diversification of R&D collaborations among partner regions. We illustrate the measure and its behaviour with respect to other conventional centrality measures by using the example of the European co-patent network at the NUTS2 level. [less ▲]

Detailed reference viewed: 66 (4 UL)
Full Text
Peer Reviewed
See detailBridging centrality as an indicator to measure the 'bridging role' of actors in networks: An application to the European Nanotechnology co-publication network
Berge, Laurent UL; Scherngell, Thomas; Wanzenböck, Iris

in Journal of Informetrics (2017), 11(4), 1031--1042

In the recent past, we can observe growing interest in STI studies in the notion of positioning indicators, shifting emphasis to actors in the innovation process and their R&D inter-linkages with other ... [more ▼]

In the recent past, we can observe growing interest in STI studies in the notion of positioning indicators, shifting emphasis to actors in the innovation process and their R&D inter-linkages with other actors. In relation to this, we suggest a new approach for assessing the positioning of actors relying on the notion of bridging centrality (BC). Based on the concept of bridging paths, i.e. a set of two links connecting three actors across three different aggregate nodes (e.g. organisations, or regions), we argue that triangulation in networks is a key issue for knowledge recombinations and the extension of an actor’s knowledge base. As bridges are most often not empirically observable at the individual level of research teams, we propose an approximated BC measure that provides a flexible framework for dealing with the aggregation problem in positioning actors. Hereby, BC is viewed as a function of an aggregate node's (i) participation intensity in the network, (ii) its openness to other nodes (i.e. the relative outward orientation of network links), and iii) the diversification of links to other nodes. In addition, we propose a generalised version of the BC measure that accounts for different node categories. An illustrative example on the European Nanotechnology co-publication network observed at the level of organisations demonstrates the usefulness and complementary interpretation power in comparison to conventional centrality measures. [less ▲]

Detailed reference viewed: 41 (1 UL)
Full Text
Peer Reviewed
See detailBridging centrality: A new indicator to measure the positioning of actors in R&D networks
Scherngell, Thomas; Wanzenböck, Iris; Berge, Laurent UL

in STI conference 2016 Valencia (2016)

Detailed reference viewed: 90 (1 UL)
Full Text
See detailCentrality of regions in R&D networks: Conceptual clarifications and a new measure
Berge, Laurent UL; Wanzenböck, Iris; Scherngell, Thomas

E-print/Working paper (2015)

This paper aims at introducing a novel measure of regional centrality in the context of R&D networks. We first demonstrate some substantial problems of SNA-based centrality measures to cope with regional ... [more ▼]

This paper aims at introducing a novel measure of regional centrality in the context of R&D networks. We first demonstrate some substantial problems of SNA-based centrality measures to cope with regional R&D networks in a meaningful way. Then, we introduce a new measurement approach of regional network centrality based on the concept of inter-regional bridging paths (indirect connections at the regional level). We show that the formal definition of our regional bridging centrality measure can be expressed in terms of three simple components: the participation intensity of a region in inter-regional R&D collaborations, the relative outward orientation in terms of all established links and the diversification of R&D collaborations among partner regions. We illustrate the measure and its behaviour with respect to other conventional centrality measures by using the European co-patent network at the NUTS 2 level. [less ▲]

Detailed reference viewed: 47 (0 UL)
Full Text
Peer Reviewed
See detailHDclassif: An R Package for Model-Based Clustering and Discriminant Analysis of High-Dimensional Data
Berge, Laurent UL; Bouveyron, Charles; Girard, Stéphane

in Journal of Statistical Software (2012), 46(6),

This paper presents the R package HDclassif which is devoted to the clustering and the discriminant analysis of high-dimensional data. The classification methods proposed in the package result from a new ... [more ▼]

This paper presents the R package HDclassif which is devoted to the clustering and the discriminant analysis of high-dimensional data. The classification methods proposed in the package result from a new parametrization of the Gaussian mixture model which combines the idea of dimension reduction and model constraints on the covariance matrices. The supervised classification method using this parametrization is called high dimensional discriminant analysis (HDDA). In a similar manner, the associated clustering method is called high dimensional data clustering (HDDC) and uses the expectation-maximization algorithm for inference. In order to correctly fit the data, both methods estimate the specific subspace and the intrinsic dimension of the groups. Due to the constraints on the covariance matrices, the number of parameters to estimate is significantly lower than other model-based methods and this allows the methods to be stable and efficient in high dimensions. Two introductory examples illustrated with R codes allow the user to discover the hdda and hddc functions. Experiments on simulated and real datasets also compare HDDC and HDDA with existing classification methods on high-dimensional datasets. HDclassif is a free software and distributed under the general public license, as part of the R software project. [less ▲]

Detailed reference viewed: 41 (3 UL)