Week 4: CCK11 data network settings

Network Import Settings

This page is the companion page to the reflective report, “Interpretation of Creative Potential in Network Dynamics across Social Media”.

1.) Network Gephi Import Settings (Input files: v1.x; Output files: 1.1: Directed; Normalization: No)

NOTE: It let me do it. The chart data was produced with network tables mistakenly loaded as directed. See below for undirected settings.

  1. twitter_cck11_graph_6weeks.net
    • Graph Type: Directed
    • # of Nodes: 498
    • # of Edges: 749
    • Auto-scale: yes
    • Create Missing Nodes: yes
  2. twitter_cck11_graph_all.net
    • Graph Type: Directed
    • # of Nodes: 835
    • # of Edges: 1188
    • Auto-scale: yes
    • Create Missing Nodes: yes
  3. blogs_cck11_graph_6weeks.net
    • Graph Type: Directed
    • # of Nodes: 158
    • # of Edges: 147
    • Auto-scale: yes
    • Create Missing Nodes: yes
  4. blogs_cck11_graph_12weeks.net
    • Graph Type: Directed
    • # of Nodes: 194
    • # of Edges: 205
    • Auto-scale: yes
    • Create Missing Nodes: yes

2) Network Measures (Output files: v1.2: Directed; Normalization: Yes;)

2.1) Centrality Measures.


  • (Statistics) > Network Overview -> Network Diameter
    • Directed
    • Normalize Centralities in (1,0): NoYes
NetworkDia
CCK_twitter_v1.2-t611
CCK_twitter_v1.2-t129
CCK_blogs_v1.2-t65
CCK_blogs_v1.2-t126

2.2) Density measures.


  • (Statistics) > Network Overview -> Graph Density
    • Directed
NetworkDensity
CCK_twitter_v1.2-t60.003
CCK_twitter_v1.2-t120.002
CCK_blogs_v1.2-t60.006
CCK_blogs_v1.2-t120.005

2.3) Degree measures.


  • (Statistics) > Network Overview -> Average Degree
    • Directed
NetworkAD
CCK_twitter_v1.2-t61.504
CCK_twitter_v1.2-t121.423
CCK_blogs_v1.2-t60.930
CCK_blogs_v1.2-t121.057

3.0) Network Measures (Input: v3.x; Normalization: Yes; Undirected;)

3.1) Centrality measures.


  • (Statistics) > Network Overview -> Network Diameter
    • Directed
    • Normalize Centralities in (1,0): yes
NetworkDia
CCK_twitter_v3.1-t128
CCK_blogs_v3.1-t127

3.2) Degree measures.


  • (Statistics) > Network Overview -> Average Degree
    • Directed
NetworkAD
CCK_twitter_v3.1-t122.711
CCK_blogs_v3.1-t122.010

3.3) Density measures.


  • (Statistics) > Network Overview -> Graph Density
    • Directed
NetworkDensity
CCK_twitter_v3.1-t120.003
CCK_blogs_v3.1-t120.010

3.4) Connected Component measures.


  • (Statistics) > Network Overview -> Connected Components
  • Directed
  • (Filters) > Topology > Giant Component
  • (Filters) > Topology > Degree Range
Networkcc
CCK_twitter_v3.1-t12187
CCK_blogs_v3.1-t1261

4) Gephi network configuration settings for image exports (Output files: 3.1).

  • CCK_xxxx_v3.1a-tn
    • (Ranking) > Nodes > Betweenness Centrality: Size 20-100
    • scale nodes starting @ 15
  • CCK_xxxx_v3.1b-tn
    • Filter > degree: 2-49
    • (Partition) > Component Id
  • CCK_xxxxx_v3.1c-tn
    • Filter > degree: off
    • (Ranking) > Nodes > Closeness Centrality: Size 10-100
  • CCK_xxxxx_v3.1d-tn
    • Filter > degree: on
    • labels: on, ID

Note on version numbering.

There appears to be a correlation between numbered items on this page and output file name numbers. This is only coincidental. Output file name numbering change with any significant change. In this example, the first batch of Gephi files (v1.x) formed the starting point of the analysis; however, fact checking phase found the original files and their description file. This explained these networks are all undirected. However at this point the first draft of the report was already made, including the histogram and scatter plot diagrams above. Moving forward, a second file batch was created from the original files and reloaded as undirected networks (v3.1). There was something else in between, but the benefit of this system is I don’t have to remember what it was, as long as I know what I’m working with and the numbers match when I place the files in the final report.

Independently, but following the same rule, analysis and manipulations of Gephi networks which are selected as figures are processed according to the report plan. This record permits individual factors to be changed and maintain the same transformations for the remaining factors between versions.

Finally, Tableau files begin with their own number scheme. Which leads to the next rule. Each application has its own series, Major number for input and Minor number for output. For example, once a directed or undirected network is imported, and basic Statistical analysis are performed, the file is saved. This state is also repeated in the Network Data Tables; therefore, Gephi files (.gephi) and Data Table exports (.csv) always begin with the same series number. Changing the Gephi import, this also changes the Data Table export series Major number (x.0). However, adding or subtracting columns in the Data Table increments the series minor number (0.x). This is for workflow reasons. It’s easier to create consistent base from on series of Gephi imports, while creating any number of export/output files as idea branches sprout or are trimmed. Similarly, images based on manipulations follow the same Major.Minor branching as table exports.

This is similar to Major.Minor Software Versioning–Sequence-Based Identifiers system [via Wikipedia].