[MP-Ryan]

Every statistical analysis is "automated" in the sense of being a computation, so I really don't get where you are going there with that word. The paper is fine and dandy, in that sense. It would only get interesting if it really predicted anything outside of its own database.

The statistical analysis was automated in the sense that a human being did not pick what things identified "male" or "female" writing.  The tool is BASED on the methodology and algorithm used in the paper - so it IS predicting outside of the database used to create it.


Read.  The.  Paper.

[Polpolion]

MP-Ryan is completely correct. If you even skim over page 5 - 6 of the paper you'll see that the data was gathered quite objectively. An examination of the source code of the tool reveals its operation to be really quite simple: It splits the input into a list of words, and then searches that list for occurences of particular words which have been determined to be useful (ie statistically viable) in distinguishing texts written by one gender from tests written by the other gender. Based on the word, it adds a certain weight to a score for either informal or formal words. Words suggesting a male author have a positive weight and words suggesting a female author have a negative weight, and the absolute value obviously denotes how much it suggests its respective gender. The algorithm the tool uses really has little to do with the algorithm used to determine the weights of each word, merely being an application of the findings of the study that used the algorithm. In fact, the tool on the website could have been written by a first year programming student.

[Dragon]

TBH, I never expected this thing to work. If anybody could guess the writer's gender based on one example of his/her writing, it'd be a human psychologist who worked on that for these 30 years, and even then, he/she won't be 100% sucessfull. I tend to be skeptical about both statistics and psychology, and this is "statistical psychology", conducted by a machine to boot. In order for computer to handle this, it'd need to have a real artificial intelligence (and even then, it'd still make mistakes from time to time).

[Polpolion]

Well if you're still skeptical of the study, you've got a tool here that can confirm its results. Even disregarding any support shown here, I'd bet money that it's right a fair amount more often than it's wrong.

[NGTM-1R]

I have to admit, I'm likewise fairly convinced that if you present it truthful data it will likely return a truthful answer.

Except about being European. That it's got a problem with.

[Scourge of Ages]

I could never stomach reading papers like this, so if somebody (Ryan) who did could sum up: what is with the European thing?
Did they find that European writers tend to use more of the same words regardless of whether they were male or female?

[Dark Hunter]

What I suspect is that the software was written to analyze American English. Europeans use the language a bit differently, so they're acknowledging that that might mess with its results.

[Droid803]

Code: [Select]

Total words: 1439

Genre: Informal
  Female = 346
  Male   = 4845
  Difference = 4499; 93.33%
  Verdict: MALE

Genre: Formal
  Female = 1048
  Male   = 2498
  Difference = 1450; 70.44%
  Verdict: MALE


Have it the first few paragraphs of a research essay I wrote for a uni class on eroge/anime.
Yeah, I did a lot of what it said males generally do. :/

Guess paper on the wonders of 2D women is manly.

[S-99]

Very interesting this gender guesser. I'm obviously not that inspired or caring to type more into it. Just wanted to see what it would do with such blatancy. I've had my fun.

[Mikes]

So i typed the word "sex" in 300 times (hey it said 300 words minimum!!) and it couldn't decide the gender... mmmh

[Pred the Penguin]

It'll basically read that as 1 word...

[Mongoose]

You could try the lyrics to "I Just Had Sex" instead.

[LordPomposity]

What you gon' do with all that junk?
All that junk inside your trunk?
I'ma get, get, get, get, you drunk,
Get you love drunk off my hump.
My hump, my hump, my hump, my hump, my hump,
My hump, my hump, my hump, my lovely little lumps (Check it out)

I drive these brothers crazy,
I do it on the daily,
They treat me really nicely,
They buy me all these ices.
Dolce & Gabbana,
Fendi and that Donna
Karan, they be sharin'
All their money got me wearin' fly
Brother I ain't askin,
They say they love my ass ‘n,
Seven Jeans, True Religion's,
I say no, but they keep givin'
So I keep on takin'
And no I ain't taken
We can keep on datin'
I keep on demonstrating.

My love (love), my love, my love, my love (love)
You love my lady lumps (love),
My hump, my hump, my hump (love),
My humps they got you,

She's got me spending.
(Oh) Spendin' all your money on me and spending time on me.
She's got me spendin'.
(Oh) Spendin' all your money on me, up on me, on me

What you gon' do with all that junk?
All that junk inside that trunk?
I'ma get, get, get, get, you drunk,
Get you love drunk off my hump.
What you gon' do with all that ass?
All that ass inside them jeans?
I'm a make, make, make, make you scream
Make you scream, make you scream.
Cos of my hump (ha), my hump, my hump, my hump (what).
My hump, my hump, my hump (ha), my lovely lady lumps (Check it out)

I met a girl down at the disco.
She said hey, hey, hey yea let's go.
I could be your baby, you can be my honey
Let's spend time not money.
I mix your milk wit my cocoa puff,
Milky, milky cocoa,
Mix your milk with my cocoa puff, milky, milky riiiiiiight.

They say I'm really sexy,
The boys they wanna sex me.
They always standing next to me,
Always dancing next to me,
Tryin' a feel my hump, hump.
Lookin' at my lump, lump.
You can look but you can't touch it,
If you touch it I'ma start some drama,
You don't want no drama,
No, no drama, no, no, no, no drama
So don't pull on my hand boy,
You ain't my man, boy,
I'm just tryn'a dance boy,
And move my hump.

My hump, my hump, my hump, my hump,
My hump, my hump, my hump, my hump, my hump, my hump.
My lovely lady lumps (lumps)
My lovely lady lumps (lumps)
My lovely lady lumps (lumps)
In the back and in the front (lumps)
My lovin' got you,

She's got me spendin'.
(Oh) Spendin' all your money on me and spending time on me.
She's got me spendin'.
(Oh) Spendin' all your money on me, up on me, on me.

What you gon' do with all that junk?
All that junk inside that trunk?
I'ma get, get, get, get you drunk,
Get you love drunk off my hump.
What you gon' do with all that ass?
All that ass inside them jeans?
I'ma make, make, make, make you scream
Make you scream, make you scream.
What you gon' do with all that junk?
All that junk inside that trunk?
I'ma get, get, get, get you drunk,
Get you love drunk off this hump.
What you gon' do wit all that breast?
All that breast inside that shirt?
I'ma make, make, make, make you work
Make you work, work, make you work.

(A-ha, a-ha, a-ha, a-ha) [x4]

She's got me spendin'.
(Oh) Spendin' all your money on me and spendin' time on me
She's got me spendin'.
(Oh) Spendin' all your money on me, up on me, on me.

Code: [Select]

Genre: Informal
  Female = 292
  Male   = 232
  Difference = -60; 44.27%
  Verdict: Weak FEMALE

Weak emphasis could indicate European.

Genre: Formal
  Female = 845
  Male   = 430
  Difference = -415; 33.72%
  Verdict: FEMALE
Maybe they're on to something.

[watsisname]

baaahahahahahaha