Hello,
I'm new to this, so apologies for possible duplication. Here's what I'm doing and where getting stuck - would appreciate help: I'm trying to 'see' (analyze) the relationships (political, economic, personal, etc.) that exist among participants in a particular public policy environment. I've only 145 vertices (i.e., participants).
I'm using 'betweenness centrality' to show the importance of each node in the network (followed suggestion in Peter Aldhous's Feb 24, 2012 paper). In the network I'm analyzing, there are 5 (five) groups of participants (e.g., corporations, NGO's, etc.). I've manually assigned each vertex to a particular group, e.g., Red Cross is assigned to the NGO group and so on. I'm using Harel-Koren fast multiscale algorithm to layout the graph for ALL participants (vertices) - the outcome shows connections and so forth. So far so good.
Now, here are my questions and where I'm stuck: (i) I'd like to analyze each of these five groups separately and get five separate graphs, but not sure how to do it without creating five separate NodeXL files, (ii) I'm having difficulty getting image that is legible when transferred for inserting into MS Word or ppt - what is the best way to do it? (iii) should I be using something other than 'betweenness centrality?' for vertex size, and (iv) should I be using something other than the Harel-Koren algorithm?
Thank you very much and please let me know if any clarification needed.
I'm new to this, so apologies for possible duplication. Here's what I'm doing and where getting stuck - would appreciate help: I'm trying to 'see' (analyze) the relationships (political, economic, personal, etc.) that exist among participants in a particular public policy environment. I've only 145 vertices (i.e., participants).
I'm using 'betweenness centrality' to show the importance of each node in the network (followed suggestion in Peter Aldhous's Feb 24, 2012 paper). In the network I'm analyzing, there are 5 (five) groups of participants (e.g., corporations, NGO's, etc.). I've manually assigned each vertex to a particular group, e.g., Red Cross is assigned to the NGO group and so on. I'm using Harel-Koren fast multiscale algorithm to layout the graph for ALL participants (vertices) - the outcome shows connections and so forth. So far so good.
Now, here are my questions and where I'm stuck: (i) I'd like to analyze each of these five groups separately and get five separate graphs, but not sure how to do it without creating five separate NodeXL files, (ii) I'm having difficulty getting image that is legible when transferred for inserting into MS Word or ppt - what is the best way to do it? (iii) should I be using something other than 'betweenness centrality?' for vertex size, and (iv) should I be using something other than the Harel-Koren algorithm?
Thank you very much and please let me know if any clarification needed.
