Data Mining for Social Network Data by Nasrullah Memon Jennifer Jie Xu David L. Hicks & Hsinchun Chen

Author:Nasrullah Memon, Jennifer Jie Xu, David L. Hicks & Hsinchun Chen
Language: eng
Format: epub
Publisher: Springer US, Boston, MA

Network collapse: Once teams are identified, the network is collapsed with each team replaced by a single composite node. Essentially, the composite “team” node is structurally equivalent [27] to the combination of all the individual nodes in that it preserves the connections between the teams members and individuals outside. Thus, in Fig. 6.3a each of the large circles containing team members would be collapsed into a single node, the resulting nodes and connections are shown in Fig. 6.3b.

Weak component identification: Once the network is collapsed, all the connected components in the network are identified and the measure is calculated for each component separately. This is different from some previous studies that use only the largest component in the network to calculate measures. This is important since different divisions of an organization may have self-contained groups of inventors and calculating measures for the largest component would ignore smaller groups.

RW betweenness calculation: Random walk betweenness (using the Newman [31] procedure for RW) scores are calculated for each node in each component of the collapsed network. Thus, the RWT measure calculated the RW betweenness score for entire teams taken as one node and single inventors who are not part of any teams. For statistical analysis, every individual in a team received the same RWT score. We believe that these new RWT scores will explain innovation diffusion better and identify individuals whose knowledge is valued for recombination within an organization.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.