If one took an academic text, a short story, a novel, a blog post, or some other more careful piece of writing, it is unclear whether this kind of classifier would work as well.
Is it? I think it's rather interesting. Though not really as much of a language dialect rather than a perspective shift, as if the men and women on average interpret the world differently.
It should be a pretty easy test to do, for those with the trained classifier; I wonder why they haven't reported on doing it.
> we noticed ‘my wife’ and ‘my girlfriend’ emerged as strongly correlated in the male results, while simply ‘husband’ and ‘boyfriend’ were most predictive for females. Investigating the frequency data revealed that males did in fact precede such references to their opposite-sex partner with ‘my’ significantly more often than females. On the other hand, females were more likely to precede ‘husband’ or ‘boyfriend’ with ‘her’ or ‘amazing’ and a greater variety of words, which is why ‘my husband’ was not more predictive than ‘husband’ alone. Furthermore, this suggests the male preference for the possessive ‘my’ is at least partially due to a lack of talking about others' partners.
To disentangle the effect of topic choice from word choice, you'd probably need a controlled study with fixed writing prompts.
"We analyzed 700 million words, phrases, and topic instances collected from the Facebook messages of 75,000 volunteers, who also took standard personality tests, and found striking variations in language with personality, gender, and age."
https://journals.plos.org/plosone/article?id=10.1371/journal...