Researchers Have Found a Way To Root Out Identity Thieves By Analyzing Their Mouse Movements With AI (qz.com) 62
An anonymous reader shares an article: In the study, published recently in PLoS One, the researchers quizzed 40 respondents about their personal details. Half of the respondents were asked to answer the questions truthfully, but the other half were given details about fake identities they had to memorize and use in the quiz.
The computer quiz kept track of the movement of each respondent's mouse as they answered the questions, and noted how the fakes differed from the truth-tellers when they moved the cursor from the bottom of the screen to the answers at the top. The quiz consisted of 12 questions like, "Do you live in Padua?" and "Are you Italian?" That covered details an identity thief could easily remember and answer, but then the quiz threw them a curve ball. "What is your zodiac sign," it asked in the second series of 12 questions, which were designed to be easy for the genuine respondents, but more difficult for the fakers to work out. After the researchers took the mouse-movement data collected from the quizzes and trained a machine-learning algorithm to analyze it, they found that was indeed the case. It was able to discern the fake responses from the real ones 95% of the time.
zodiac signs (Score:4, Insightful)
I have no fucking clue what my zodiac sign is, I'd have to look that crap up. I'm not just not into fairy tales for adults.
Re:zodiac signs (Score:5, Insightful)
I have no fucking clue what my zodiac sign is, I'd have to look that crap up. I'm not just not into fairy tales for adults.
My guess is that makes you part of the minority. I, like I'd imagine most people, know my sign but don't particularly care. I have no idea what it "says about me" and put zero faith in sign-specific predictions. I know what Chinese-animal-year I was born in too, but put the same amount of faith in what it implies.
Just because you don't believe in Hansel & Gretel doesn't mean you've never heard of it. You've really never run into an astrological chart?
Re: (Score:2)
Of course I've seen an astrological chart, that doesn't mean I have the damn thing memorized either. I've also seen a bible as well, doesn't mean I could quote anything from it or believe anything from it either.
Can you name one of the books from the Bible? If there was one specifically assigned to you for some reason, might it stick in memory? Knowing your sign doesn't imply memorizing the whole chart nor believing in it. It involves remembering one of 12 signs. Most of us, I'd imagine, remember it entirely unintentionally.
Re:zodiac signs (Score:5, Funny)
Well, you've got your Fellowship, Two Towers and then Return of the King.
What do I get?
Re: (Score:2)
Yeah, nobody said anything about memorizing an entire astrology chart. Who's putting words in whose mouth again?
Re: (Score:3)
This actually points out how culturally-specific the question is. I have no idea what my Chinese animal year is, but the Chinese version of this would no doubt ask me. Immigrants are going to be automatically labeled as identity thieves by this algorithm because they don't share the same cultural context.
it's a craappy study. (Score:2)
someone calling you while filling the questionnaire would also mark you as an id thief.
using a touchscreen would do that.
answering a message in the middle of filling it would also mark you as an id thief.
the whole point of the study is this: if you need to look up something you're likely to not be able to answer as quick and straightforward.
it's the kind of a study you can make a hypothesis for easily and verify it easily and provide something that can be presented like it is a solution possibly with furthe
Re: (Score:2)
I have no fucking clue what my zodiac sign is
I don't believe you. If you have this strong of an opinion about astrology, then you have almost certainly been exposed to it enough to know your sign. Astrology is idiotic nonsense, but I have known my sign as long as I can remember. I have had plenty of people tell me my sign and read my horoscope out loud. If you grew up in America, and associated at all with "normal" people, I don't see how you could avoid it.
Re: zodiac signs (Score:2)
I think it is maybe a generational thing. I have no issues believing they don't know their sign. When I was younger, I believe I'm a bit older than you, it was pretty common to ask what someone's sign was. We also had some fruitcakes who believed in it.
I'd be surprised to find out that both of my kids know their sign. I suspect my daughter would, she is in her thirties. My son? Probably not. He's a bit younger than she. I'm pretty sure they'd both know their Chinese animal sign. I was the custodial parent.
Re: (Score:2)
I saw an underage guy get blocked from a bar based on not knowing his zodiac sign. He'd borrowed the ID of a slightly older friend who looked a bit like him (but not enough). He'd learned by heart everything on the ID card, but the bouncer got suspicious and then asked him for his zodiac sign.
Maybe a generational thing indeed, this was some twenty years ago.
Re: (Score:3)
This Just In: Data Scientists develop a machine-learning algorithm that can identify, with 95.6% accuracy, people who fake not knowing their Zodiac sign, based on how vitriolic their slashdot comments are.
Respondents were shown an article on slashdot regarding a supervised AI application, and encouraged to comment. The forum then threw them a curve-ball, encouraging them to disclose their star signs. The results were then analyzed by a machine learning algorithm.
Johan Virgo, when asked about the research
Re: (Score:2)
This just in: Data Scientists develop a machine learning algorithm that can detect people, with 95.6% accuracy, who fake not knowing their zodiac sign, based on how vitriolic their slashdot comments are.
Participants were shown a Slashdot article and encouraged to comment. The process then threw them a curve-ball, allowing them to respond as anonymous cowards. Finally the participants were asked to disclose their Zodiac signs. The results were then fed in to a supervised AI.
Johan Virgo, when asked about the
Re: (Score:2)
Re: (Score:2)
Re: (Score:3)
Then you're clearly a Sagittarius.
Re: (Score:2)
I might know my star sign if the people who publish such things could figure it out. I was born right at the end of one of the signs, and it varies, depending who wrote out the fairytales that day. This must apply to 12/365 of people. It sounds like a really stupid question - nearly as bad as What Chinese Year were you born in? Goat, Dragon, which? I think a high proportion of folk would not know.
How tall are you? - sounds like a much better question. I somehow doubt most identity thieves would actually kno
What is my sign? (Score:4, Informative)
I respond .... "Neon".
Also works for signature pads (Score:2)
This can also work with signature pads. A forger maybe able to fake your signature, but they have a much harder time faking the rhythm and timing of the strokes.
Another test that measures timing is testing for pedophila. Patients are shown pictures of children, adolescents, and adults, and asked to rate each picture for attraction. The ratings are then ignored, and instead the "dwell time" for each photo is what is actually measured, since that is much more predictive.
Re:Also works for signature pads (Score:4)
Why would there be a test for pedophilia? Is that for use in a clinical setting? I know some issues are ... tricky for people to deal with. Self-denial and all. I'm not 100% certain how detection is helpful, though, considering a person who is aware of some issue and willing to accept the potential shouldn't have trouble introspecting and identifying that they maybe are kind-of bisexual or whatever it is they're choking on.
I'd imagine test for pedophilia would be ... problematic in other situations. Huge invasion of privacy, at least here in the U.S.. We can't take legal action against someone who hasn't taken actual criminal action (no thought-crime), and discrimination against a person for such a diagnostic result would be devastating. We have legal protections for discrimination and privacy invasion if you're gay or something.
Re: (Score:1)
You should be careful, because in today's world (especially in the US), not being hasty in throwing people under the bus is, in itself, a proof of self guilt.
Re: (Score:2)
Psh. I have no friends and am schizoid, and people have tried to use this in political debates about welfare to argue that a sociopath obviously has no valid political opinion on social welfare (I literally score 0 on the sociopath test some days; the threshold is a mile and a half away).
Oddly enough, being tossed under the bus hasn't had an impact on me; it's just plain old not having a need to mesh into a social group that makes me question things. You could argue, in an abstract way, that maybe you
Re: (Score:2)
Why would there be a test for pedophilia? Is that for use in a clinical setting?
Where else would you test for a medical condition than in a clinical setting?
Re: Also works for signature pads (Score:2)
Condition of release, testing prior to sentencing, and determination of associated risks when being considered for release from mandatory treatment - to name a few reasons that I can think of. They do another one where they put a cuff, like for blood pressure, on your penis. There are several documentaries that cover the subject.
Re: (Score:2)
If they have confirmation of pedophilia, that's pretty much it. You can't un-pedophile someone any more than you can un-gay them; all you can do is determine if they're liable to tie a cookie to a string and troll it through the nearest preschool yard.
Re: (Score:2)
"Researcher" ? (Score:2)
Re:"Researcher" ? (Score:4, Interesting)
AI versus AI (Score:2)
Re: (Score:2)
Then to fool the AI, all you need is an AI program to drive the mouse to mimic the original person.
Generative Adversarial Networks [wikipedia.org]
Claims are overblown (Score:4, Interesting)
By itself, this study just shows that a machine learning algorithm can compute a statistical average (regarding mouse movements) that classifies a specific set of 40 people into two groups.
If the same statistical average also classifies other groups of people accurately, then you can make real claims about separating truth-tellers and liars.
See this good explanation of pattern classification [lisafeldmanbarrett.com] and its uses and misuses.
Re: (Score:2)
They also had to memorize answers. If they had them on another open window, there might be odd mouse movements; if they had them on another monitor, then they would have less to think about. Everyone has to read all the answers and compare with their own memory; struggling with recall is potentially-detectable, and this may only show how well someone learned something. If the data isn't memorized at all, are the artifacts as large?
OMG! I stole my own identity! (Score:2)
Worse, now that this technique has been revealed, the real thieves meanwhile will remember to note down info they enter in pieces of paper as the go along, thus they can revisit, open another google window to search for stuff etc.
They've found nothing (Score:2)
But their dowsing rod is now better than ever... or was in a test, anyway.
40 participants (Score:2)
No significant result possible.
This May Prove To Be Useful Actually (Score:2)
ID thieves don't use mice (Score:2)
GOOGLE (Score:2)
At around 2012 - 2014 Google's Russian office developed a very similar technology for clickfraud detection. What has happened later - the best talent involved in developing it was given a relocation offer to Switzerland given that he was just SO INCREDIBLY VALUABLE, but the guy refused. Their HR's became more assertive, threatening to ruin his Arbeitsbuch (in Russia your work record is an official document just like in China). The guy demandes his Arbeitsbuch back and resigned the same day. People say he j
Re: (Score:2)
Coincidentally you have to run javascript (Score:2)
Oh look, another reason why a web form has to have javascript to work. It's for security, doncha know!
Sounds like straightforward classification (Score:2)
Interesting but not novel or new. For example typing patterns have been used before, this is an established biometric measurement called typing or keystroke dynamics [wikipedia.org] . There are even companies [typingdna.com] that sell typing metrics as an authentication factor.