[SOLVED] Some truly n00b question(s)

[rd0002]

Hi there, and thanks for a beautiful app !

I have a data table with many records which represents in my view "nodes" (the balls or eggs which are connected by lines).
Each record (they represent organizations) has a NAME and several other attributes like a person, address, some money amounts, ...
I SUSPECT that connections exist between these organizations. These connections are commonality of NAME, a person, an address, ...
These common values of certain attributes would be the EDGES of the network.
My records look like this
"ORG_NAME";"ORG_CAREOF";"ORG_CITY";"ORG_ZIPCODE";"income";"revenue"

My questions:
- What must I do to load this data and let gephi find out the connections between these records?
- Is this something gephi is able to do or must I use something else before giving the data to gephi?
- Would it be better to format my data in another way before giving it to gephi?

Thanks again.

[seinecle]

nOOb answers:

I don't know of any standard way to do it. Maybe that UCINET (a very common network analysis program) offers it?

Network WorkBench does it automatically for bibliographic data (you feed a list of academic papers in the approriate bibliographic format, and it returns a network of co-authors of these papers). But this demands quite some time to get acquainted to all the issues in formatting your data correctly, importing it in NWB, and re-exporting it in a form useable by GEPHI or any other software.

Another way that I use for social network data is starting from scratch on Excel, but that requires some knowledge in formulas and macros.

I'd be interested in the solutions used by other members of this forum!

Best,

Clement

[admin]

Hi,

The Data Laboratory will let you import your Excel file and select the appropriate columns to match nodes and edges: http://gephi.org/2010/introducing-data-laboratory/

It will be available for testing purpose in Gephi 0.8 alpha, but no date has been fixed.