This page is referenced by:

• 1 2017-01-21T15:31:30-08:00 Section 2: Network Analysis / ANALYSIS 13 Methods and analysis SECTION 2 (Network Analysis): Social Media Analysis + Strong Connectivity (most connected) + Content analysis + Likes per country plain 2017-03-23T14:33:31-07:00

  Methods and analysis

  • Social Media Analysis
  • Strong Connectivity (most connected)
  • Content analysis
  • Likes per country

  The second part of the analysis corresponds to the social media and network observations for the two selected hubs and analysis of the global presence of the totally of companies with facebook pages using data containing information on likes per country of each Facebook page, to download this data I used the Netvizz Facebook app. Firsts results after checking for the online presence of all indexed companies after data validation show that there is not a prevalent use of social media platforms (38% of companies with Facebook pages and 45% of companies with Twitter accounts) and that there is a symbolic global representation of interaction points outside the UK.
  
  The combination of data from posts by page and users downloaded from the C4DI Facebook page (in red) and The Enterprise Centre Facebook pages (in black) using Netvizz and imported into Gephi as append graphs show the centrality of the C4DI network after applying Force Atlas 2 layout. Other calculations of strongest connectivity applied to the node size help visualize the most connected items in order to perform first interaction analysis and afterward content analysis. Figure 29 show labels representing the count per like of each central node, this indicator has been used to located via data laboratory in Gephi the central node in order to perform content analysis and identify emergent patterns among the most connected items, such as that while C4D1 has a strongly connected network with an even distribution of users and posts the graph also shows evidence that point out The Enterprise Centre network as a superuser network with a fewer connectivity among posts.

  Figure 29 - Representation of strongly connected components

  Facebook network *C4DI & Enterprise Centre facebook pages

  GRAPH COMPONENTS INFO

  • Nodes: 871
  • Edges 1344
  • Directed Graph
  • Combination of data from posts by page and users downloaded from C4DI (red) + Enterprise center (black)  facebook pages using Netvizz and imported into gephi as append graphs
  • Data Timeframe  2010/08/28 - 2016/09/03
  • Data: Likes per post/user

  INTERVENTIONS

  • Node size represents strongly connected id
  • Lavel show like count of central nodes
  • Layout: Force atlas 2 + Yifan Hu
  • Filtered peripheric nodes (not connected, with no links)

  STATS

  • Average degree 3.086
  • Network Diameter 1
  • Modularity 0,619
  • Connected Components 104
  • Average path length 1