Twitter's new favourite words

[Hugo]

I wrote a Python script that searches Twitter for tweets containing "[X] is my new favourite word". It then takes all those new favourite words, logs them, lowercases them, and adds them to a list of words on Wordnik. And, from this week, it also tweets them and makes word clouds.

It also does the same for "[X] is my new favorite word" and "[X] is my new fave word" so you can see some geographic variation.

It's been going six months since 24th February 2013, runs nearly every day, and has found some 18,360 new favourite words.

Here's top tens for the whole set and each subset.

The whole data set contains 18,360 words and 10,018 unique words.
The top 10 words are:
1. no (151)
2. cunt (120)
3. sassy (99)
4. bitch (95)
5. ratchet (87)
6. fab (86)
7. twat (73)
8. cheeky (72)
9. faggot (66)
10. fuck (62)

"New favorite words" contains 10,299 words and 5,964 unique words.
The top 10 words are:
1. no (129)
2. cunt (75)
3. bitch (73)
4. ratchet (55)
5. thot (53)
6. fuck (47)
7. twat (41)
8. sassy (38)
9. fuckery (36)
10. whatever (35)

"New favourite words" contains 6,623 words and 4,505 unique words.
The top 10 words are:
1. fab (50)
2. sassy (50)
3. cunt (41)
4. cretin (36)
5. cheeky (34)
6. faggot (33)
7. fabulous (32)
8. twat (29)
9. ratchet (24)
10. serendipity (20)

"New fave words" contains 1,436 words and 1,162 unique words.
The top 10 words are:
1. fab (19)
2. sassy (11)
3. ratchet (8)
4. cheeky (7)
5. swaggot (7)
6. cretin (7)
7. coont (6)
8. bastard (6)
9. dope (6)
10. dude (5)

What does this tell us? I don't know, but whilst the top ten contains lots of swearing, around half (or about 2/3 for favourites) of each list contains unique words. Browsing the lists shows lots of neologisms (and neoswearisms).

The word lists on Wordnik:
http://www.wordnik.com/lists/twitter-favourites
http://www.wordnik.com/lists/twitter-favorites
http://www.wordnik.com/lists/twitter-faves

Follow along:
https://twitter.com/favibot

See word clouds:
http://www.flickr.com/photos/hugovk/sets/72157636928894765/

See the script:
https://github.com/hugovk/word-tools/blob/master/new_favourite_words.py

Inspired by:
http://www.wordnik.com/lists/outcasts

Hugo

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org