Prepare data for network analysis: Is it possible to use network.py with twarc2 downloaded tweets?

I’ve downloaded with twarc2 a set of tweets in .jsonl and I am now trying to create .gexf or other network usable files (list of nodes and edges being able to select which relationship to use).

While running utils/network.py it throws some errors, as the names of the variables (that’s my guess) are not the same with the API 2 (id instead of id_str, author instead of user…) and the script is unable to process them.

Which way do you recommend to transform the data into files usable for data analysis? Thanks!

[numeroteca]

I just added the minor changes of name variable and parsing noted above. See the current script: https://gist.github.com/numeroteca/aa040b0488c914d1e4a37e40117ef062

[edsu]

Ok I’ve released a port of the old network.py script as a twarc2 plugin. You should be able to install it with:

pip install twarc-network

and then run it to generate a network as HTML D3:

twarc2 network tweets.jsonl network.html

More details about the various format options are available at https://github.com/docnow/twarc-network

Please ask questions about the plugin over in that issue tracker if you don’t mind too much!