More Fun with New Media

In a prior post, we saw how Twitter Tweets can collectively be used as a mood ring for the U.S..  Well, it turns out that Facebook status updates show the times of year that you are most likely to be dumped by your significant other!  Here is the chart that shows volume of break-ups over time:

Most of this makes sense.   Funny that Mondays in April have mini-peaks (why April?).  I've heard that in the corporate world, firms usually lay off employees on Fridays.  I guess that in each case the breaker-upper chooses a time at the end of a period of relationship interaction, rather before a period of relationship interaction... but maybe that's over thinking things a bit.

I guess there are a lot of returned Christmas gifts... so there is an advantage to last minute shopping?  There is a small uptick before Valentines Day too, but people seem to stick it out better.  Until Spring break, of course.  Interesting stuff.

Check out the author's TED talk as well.  You have seen TED, haven't you?

What New Media can tell us about ourselves

In July, an interesting article made the news rounds.  A team at Northeastern University analyzed three years' of public tweets, and categorized them as "positive" or "negative" by identifying indicative words.  Then with this data, the team could measure the "mood" of the United States across different dimensions (e.g., typical mood curve throughout the day, East vs West Coast, etc.):

I admire people that take new sources of data and look at it in interesting ways like this.  Nerd envy.  So in a moment of boredom I tried something:  comparing the count of results that Google fineds for specific words.  Here's what we find:

Love:     1,080,000,000
Like:      1,480,000,000
Hate:        127,000,000

Ours:          29,000,000
Yours:         89,500,000
Mine:        184,000,000

Debt:          87,200,000
Profit:       130,000,000

Peace:      180,000,000
War:         478,000,000

Gun:          122,000,000
Kiss:          135,000,000

Fun:          443,000,000
Work:    1,180,000,000

Sour:           18,500,000
Bitter:          30,100,000
Sweet:       212,000,000

Candy:        77,600,000
Healthy:    131,000,000

Earth:        279,000,000
Water:       590,000,000
Air:             661,000,000
Fire:        2,290,000,000

Yes, I know that I'm not controlling for homonyms, but it was somewhat fun and interesting to do.  I was surprised at so many results for "fire" -- I had thought that "earth" would have more results.  readers:  any thoughts from your perspective?