I just found this little gem buried on the last page of the June/July issue of Geek Monthly: the Gender Genie. Just paste in something by your favorite novelist — or an excerpt from a newspaper article or your own blog — and the Gender Genie will guess the gender of the author. (For best results, the text should be more than 500 words.) The Gender Genie is based on an Illinois Institute of Technology study that explored the "lexical and syntactic features" of fiction and nonfiction.
I had to try to confuse the Genie, of course. Maybe it can predict the easy ones like Hemingway and Austen, but what about queer authors? I tried Henry James, and the Genie said “female” — aha! Then I tried Virginia Woolf and the Genie said "male" — fooled again!
Admittedly, both of those were close (the Genie gives you the female score and male score for each excerpt, and the numbers on James and Woolf weren’t far apart). So then I tried Jeanette Winterson and the Genie was way off base, even though I used a column excerpt that seemed to include traditionally "female" words like sunburst, flower and cat and was generally quite personal and introspective.
Are queer authors always gender outlaws when they write? I finally pasted in some of my own stuff. It seems that in articles and more “serious” writing, my style is male, but when recapping, I come across as overwhelmingly female. I’m blaming Simone Lahbib‘s tongue and Jennifer Beals‘ tank tops.
According to the study, a few pronouns here and there can make a big difference:
In particular, we find significant differences between
male- and female-authored documents in the use of pronouns and certain types of noun
modifiers: although the total number of nominals used by male and female authors is virtually
identical, females use many more pronouns and males use many more noun specifiers. More
generally, it is found that even in formal writing, female writing exhibits greater usage of
features identified by previous researchers as “involved” while male writing exhibits greater
usage of features which have been identified as “informational.”
Can you stump the Genie? (For fiction excerpts, try Project Gutenberg.)