That's nothing. It got Derpy Hooves in 12 questions!McCa said:
That's because it is designed to guess people and characters and whatnot. Of course it isn't going to get a car when it asks questions like "Is your character female?"TestECull said:
Akinator, the web genius who's developer must've been one very bored guy.
- Thread starter Seraj
- Start date
Haha, wow.
This thing has a large database.
It even got A Scanner Darkly's Robert Arctor! Cool!
-Tabs<3-
This thing has a large database.
It even got A Scanner Darkly's Robert Arctor! Cool!
-Tabs<3-
It almost certainly builds Hidden Markov Models (Wikipedia has a decent, though perhaps overly technical article if you're interested) for the transition probability of each set of answers to a particular target. In such a case, all it needs is a database of target nouns and a large database of associated descriptors (question answers) to build the statistical models. It could index certain areas of Wikipedia for nouns and simply model the nouns as a function of descriptors in the articles (it would be especially easy to extract article groupings and standard sidebar information for certain sections) or even in Google searches for the term. At each step, it takes the path with the highest probability of yielding a definitive transition probability, which it eventually returns as the answer.
Alternatively, because I'm not sure how it was when it started, they got a huge database of targets, created a stock set of questions, and then presented them approximately at random to the first users and modeled transition probabilities from that with each subsequent user refining the model (in other words, the system self-optimizes). It's also very possible that they're using both models and adding their transition probabilities in some way.
The thing that often makes such systems so neat is that you can completely automate model building and, given a large enough data pool either from spidering or crowdsourcing, you can build extremely robust models that often prove surprising accurate to non-machine minds.
Alternatively, because I'm not sure how it was when it started, they got a huge database of targets, created a stock set of questions, and then presented them approximately at random to the first users and modeled transition probabilities from that with each subsequent user refining the model (in other words, the system self-optimizes). It's also very possible that they're using both models and adding their transition probabilities in some way.
The thing that often makes such systems so neat is that you can completely automate model building and, given a large enough data pool either from spidering or crowdsourcing, you can build extremely robust models that often prove surprising accurate to non-machine minds.
Oh hohoho!Macgyvercas said:
You sly devil. Wish I'd thought of that >.>
Now it knows my secrets! eep!
-Tabs<3-
tryed using the character super meat boy.. it failed, guessed L-block from tetris.. ah no you can continue, it actually got super meat boy on it's 30th question.
also got gintoki from gintama pretty fast.. kaiji in 21.. gonna throw him a harder one now..
hmm it also got arsene lupin at around 30 ish..
wow were gonna use a norwegian person and it lucked out and asked if it was a norwegian character in the third question.. err ok this is kinda wierd it also after that asked me if he was european, and then if he was from the scandinavian area.. (which are both places norway is in..) and now it asked me if it's swedish? wtf.. also it sugested "reodor felgen" after i answered he was not associated with cars, which that character is in a huge way. also asking me if he is an animal after i've answered he was human.. and it's asked me if the character is a father and if it got children a couple times.. oh my holy shit balls on 30 it got it.. it was "kjell" from a random norwegian comic which in the past year probably have gained some atention. that's pretty sick. : P (someone have played him before though so that may be the only reason..)
rofl was going with another norwegian cartoon char and it guessed "kjell" on the 12th question, wrongly tho, and wierdly since this time i answered it was someone with a kid.. well defeated him with this one.
viggo from the cartoon magazine with the same name. it's an older cartoon.
also got gintoki from gintama pretty fast.. kaiji in 21.. gonna throw him a harder one now..
hmm it also got arsene lupin at around 30 ish..
wow were gonna use a norwegian person and it lucked out and asked if it was a norwegian character in the third question.. err ok this is kinda wierd it also after that asked me if he was european, and then if he was from the scandinavian area.. (which are both places norway is in..) and now it asked me if it's swedish? wtf.. also it sugested "reodor felgen" after i answered he was not associated with cars, which that character is in a huge way. also asking me if he is an animal after i've answered he was human.. and it's asked me if the character is a father and if it got children a couple times.. oh my holy shit balls on 30 it got it.. it was "kjell" from a random norwegian comic which in the past year probably have gained some atention. that's pretty sick. : P (someone have played him before though so that may be the only reason..)
rofl was going with another norwegian cartoon char and it guessed "kjell" on the 12th question, wrongly tho, and wierdly since this time i answered it was someone with a kid.. well defeated him with this one.
viggo from the cartoon magazine with the same name. it's an older cartoon.
Alright, to sum up so far, he correctly guessed:
Robin Hood
Giygas
FACE (Yume Nikki)
Blaine (Glee)
Edward "Spoony Bard" (FF4)
Clive (LittleBigPlanet 2)
Sheldon Cooper (Big Bang Theory)
The Katayanagi Twins
Emil (Nier)
Boney (Mother 3) Though at first he guessed the "Duck Hunt" dog
King (Cave Story)
Drowned Ophelia (Brutal Legend)
Luka "Cheshire" (Bayonetta)
Pascal (Tangled)
Caterpie
Derpy Hooves
Charlie Nash (I knew he knew when he asked if the character was dead and I said "I don't know")
Denmark (Hetalia)
And he didn't get:
Tepeu (Majin & The Forsaken Kingdom)
Bebedora (Arc The Lad)
Ruffgar "Ruff" Brookback (Redwall's "Lord Brocktree")
Arty (De Blob)
Burger & Fries (Angry Youtube cat)
Livane (Folklore)
Alright, I gotta stop this.
Robin Hood
Giygas
FACE (Yume Nikki)
Blaine (Glee)
Edward "Spoony Bard" (FF4)
Clive (LittleBigPlanet 2)
Sheldon Cooper (Big Bang Theory)
The Katayanagi Twins
Emil (Nier)
Boney (Mother 3) Though at first he guessed the "Duck Hunt" dog
King (Cave Story)
Drowned Ophelia (Brutal Legend)
Luka "Cheshire" (Bayonetta)
Pascal (Tangled)
Caterpie
Derpy Hooves
Charlie Nash (I knew he knew when he asked if the character was dead and I said "I don't know")
Denmark (Hetalia)
And he didn't get:
Tepeu (Majin & The Forsaken Kingdom)
Bebedora (Arc The Lad)
Ruffgar "Ruff" Brookback (Redwall's "Lord Brocktree")
Arty (De Blob)
Burger & Fries (Angry Youtube cat)
Livane (Folklore)
Alright, I gotta stop this.
after running trough a few character i've noticed it's got some flaws, even after asking if it's a woman, and i say no and then asking if it's an adult man to which i say yes it still comes out with a woman as a answer.. well in the end after 30 questions it still pulled trough and answered charlie kelly from it's always sunny in philadelphia..
I beat him using the Old World Blues Toaster and Chief Hanlon of Fallout: New Vegas! Fun little distraction, little animated genie knows his dictators and their misteresses.
