Random af project idea that will see me burned alive by the internet (because if I do it I intend to put it in dev.to which is full of "that offends me" people):

Generate a classifier that will scan text from different websites and categorize where the person might be from.

Example: "plz send bob and vagene" <--- we all know

"mami que ricas nalgas" <--- Mexican for the most part.

"there, their, they're and similar text" <--- my fellow Americans for the most part....

"cyka blyat" <--- 0.o we know

"pompous statement about the way Americans do shit" <--- European, meaning, from Yurop.

"angry as fuck rant/banter" <-- German

"lol whatever Trump is the best president ever" <--- some moron from the south of the U.S (south much like myself but I am not a Trump supporter nor a republican)

etc etc.

What makes this complex is that I would have to put together my own dataset in the highly likely chance of something like that not existing already for me to use.

Can you imagine the chaos?

  • 5
    just extract the text and plug it into a pre-trained classifier or a simple pattern matching script if you're against that ML shit. definitely a good troll. maybe it can also try generating a "personality profile" the way they do with horoscopes but add a bit of not so subtle trolling here and there

    "By the way you type you are educated and well read and you have excellent comprehension skills that combined with your above the average empathy can make you a very successful person. But by your choice of words you seem to jack off too often to certain types porn that cause irreparable damage to certain parts of your brain."
  • 2
    Yeah, the biggest initial challenge with any kind of supervised learning thing is finding the dataset. On a personal / hobby level, creating the amount of data you'd need for this kind of thing would be completely out of reach - and that's even assuming your assumptions are correct to begin with :-)
  • 7
    I feel excluded.

    *publishes Medium article*
  • 1
    @molaram Now I feel included. :)
  • 2
    @AlmondSauce it really is out of reach, I mean, technically no (you get what I mean) but in all practical terms it would be a (quoting Jian Yang) "that is a very boring work"
  • 3
    @molaram I like building my own classifiers and love ML :D
  • 1
    *Scan Trump's twitter* Turns out he is half Hispanic or Chinese. 🤣🤣 That would be hilarious.
  • 2
    @Bybit260 ...and that kids is why mexico is now a radioactive desert
  • 1
    @molaram Finds out he is half Mexican, nukes Mexico. Trump's logic. Amazing.
  • 2
    @Bybit260 MMGA Make Metsico Great Again.
  • 1
    @M1sf3t can't shake the feeling at least some of these states were more like "nay hillary" rather than "yay trump"

    not saying he won fair and square, I keep getting a vibe that elections are mostly rigged
  • 0
    This rant right here, officer
  • 0
    @uyouthe I plead not guilty your honor
  • 1
    Dating sites might be a good place to data that is tied to a location.

    Of course, people do move. :-)
  • 1
    Me to random person online "Where are you from ?"

    Random person "Oh I'm from X, my name is Y, and I like Z, here is a video of me"

    5 years later I bump into them wearing an Anonymous mask, and I say "Hello Y!"

    Y "How do you know my name !"

    Me "You are wearing the exact same clothes you wore in that video 5 years ago and sound the same, and this location is near where you said you live..
  • 2
    @Nanos made me think of my friends from different parts here in the U.S for some reason:

    Friend showing me a picture of his family
    "This is me (John Stevens), and this is my mom Clara Rogers, this is my dad Mark Hamsworth, this is my sister Clara Jackson OH and this is grandma, Petunia Willsbourgh"

    Me: "Are y'all adopted? Not one single person here has the same lastname" 0.o

    I have seen this shit so many fucking times and it always trip me the fuck out
  • 0
    @M1sf3t so, the whole map is read, yet you don't entirely agree with the whole "southern/trump" supporter thing? We might look at graphs in a different way.
  • 0
    @M1sf3t right, but I didn't consider the rest of the states, I mentioned southern for which historically it has been red, i went with the default because it's what makes sense.
Add Comment