Boo-Hooray

by Carson Reynolds

I’ve been following up on this idea for an ethical evaluation. The Cambridge philosopher A. J. Ayer was known for something called the boo-hooray theory.

So I wrote up a little prototype boo-hooray search engine. It googles for a word, then downloads all of the returned links. These are stripped of html and processed for word-frequency. Then taking a list of English words with emotional orientation (see page 13 of Cowie et al.) I find the “Boo” words and “Hooray” words.

An example query: emotemail

Boo			Hooray
----------------------------------------------
critical        21	content		102
bored   	9	happy		30
confused        7	interested      28
surprised       4	ready		19
angry   	3	patient		9
disappointed    3	satisfied       2
rejected        3	accepting       1
annoyed 	2	agreeable       1
bitter  	2	calm    	1
contrary        2	pleased		1
irritated       2	sympathetic     1
suspicious      2
ambivalent      1
awed    	1
disgusted       1
jealous 	1
uncertain       1

A “cheer” metric might be Sum(Hooray Words) / Sum(Boo Words). In the example above “cheer” = 3. If this is less than 1, then the cheer in a boo.

Another example: imwatching

Boo			Hooray
----------------------------------------------
bored		18	content		134
surprised	17	happy		43
confused	15	ready		26
critical	12	interested	25
angry		11 	patient		12
jealous		7	calm		7
disappointed	6	delighted	6
rejected	5	pleased		6
suspicious	3	satisfied	3
annoyed		2	amused		2
bitter		2	sympathetic	1
amazed		1	tolerant	1
ambivalent	1
contrary	1
discouraged	1
envious		1
irritated	1
puzzled		1
revolted	1
stubborn	1
uncertain	1

Here “cheer” = 2.46

There are some current problems: Some of the words on my “Boo” list have to do with what Plutchik calls “awe.” This really ought to be a web form so people can try random queries. The results are only limited to English words. If someone says “not” in front of one of the words, boo-hooray doesn’t detect this. “Content” is also a very common html element, and I think my sgml-stripper might not be robust enough.

Advertisements