As it is my primary means of communication, I’ve been interested in graphing my eMail correspondence to find emergent patterns and relationships. It’s my intention to take this exploration far further and deeper into linguistic and text analysis in the future (and, once I have a similar amount of data, applying the same tools to Twitter) but for now here’s a very rough start.
The corpus used is correspondence to and from my gMail account, a total of 7823 messages from July 2006 to today, with the lower two graphs showing activity for the last complete year for which data is available (2010) with identities anonymized.
So far there are no huge surprises: I did have the wry observation that I send a lot of eMail after midnight.
I used Python with mail-trends for the analysis, and Cheetah to template the graphs. Obviously all identities other than my own have been anonymized.