Timeu.se is a new a web tool that provides reports of how people spend their time based on what they tweet about.
"Some people say that Twitter conversations are 'pointless' or 'babble' or that people tweet about 'what they had for breakfast," creator Scott Golder, a sociology graduate student at Cornell, writes on the site. "But if you think about it, most casual conversations are just as inconsequential — we just don't record them forever on the web!"
LAUNCH reached out to Scott and will update this story when we hear back.
To write a query, enter a one or two-word phrase in the text box, like "breakfast" or "email address." You can also combine phrases using a plus sign. After typing a query, choose a plot type and click "Submit Query."
There are four kinds of plots: Week plot (graphs a single series over one week), Line plot (graphs up to five series, each as separate line), Scatterplot (compares activity of two series across a diagonal line) and Heat Calendar (graphs a single series on a calendar-like grid).
The dataset comprises 2.4M user accounts and more than 500M tweets that Scott and his classmate Michael Macy collected and aggregated for an academic paper, "Diurnal and Seasonal Mood Vary with Work, Sleep and Daylength Across Diverse Cultures," in early 2010. Their paper was recently published in Science.
The above scatterplot shows a correlation of the tweeting patterns of "beer" vs "drunk." According to the plot, it takes about seven hours to get drunk.
The "happy hour" Heat Calendar plot reveals that Fridays from 3-6pm are the happiest of hours for workers.
The "traffic" query shows trends Sunday through Saturday on the Week plot. Based on the plot, the worst traffic is Friday night -- assuming that people are tweeting about their traffic frustrations.
The line plot for query words "bacon, sausage" shows that bacon is more popular than sausage.
CONTACTS & LINKS
Email: sag262 at cornell dot edu