In developing a brand on Twitter (and social media in general), how does what you say and how you say it correspond to positive results (more followers, for example)?
/exp/acarrell/twitter_brand/aggregate_out/ contains files of counts, IDs, etc. aggregated from jobs run on each month in the 2017 Twitter 1% sample.
In that top directory there are .txt files for some manually selected artist roles (dancer, artist, writer, etc.) that contain IDs with descriptions containing said role.
In archive/, there are the following types of files:
{role}_agg_count.p: pickled counts users identifying as {role} (If I remember correctly, I did not ensure that IDs weren't double counted across months, just within months)
{role}.txt: from brief manual inspection, these are junk files
{ngram}_kv.tsv: ngram "i am a" counts in descending order (so highest on top)
imaout{ngram}.p: pickled versions of the files above, but unsorted (not sure what the point of these was)
@abenton :
/exp/acarrell/twitter_brand/aggregate_out/ contains files of counts, IDs, etc. aggregated from jobs run on each month in the 2017 Twitter 1% sample.
In that top directory there are .txt files for some manually selected artist roles (dancer, artist, writer, etc.) that contain IDs with descriptions containing said role.
In archive/, there are the following types of files: