Slashdot Mirror
Python Scripting and Analyzing Your Way To Love
fiannaFailMan writes "Wired reports one mathematician's mission to find love online by data mining from OK Cupid and applying mathematical modeling to optimize his profile(s). His methods included using 'Python scripts to riffle through hundreds of OkCupid survey questions. He then sorted female daters into seven clusters, like "Diverse" and "Mindful," each with distinct characteristics.' But the real work began when he started going on dates."
Slashdotter has vivid coding dream in Mom's basement and find love.
I mean, I'm glad it can do almost anything, but I'm still waiting for import antigravity to work properly.
I am officially gone from
Well, there's the step I never get to.
So the real thing here is that someone needs to be building a dating website for nerds (assuming it's not already happened).
Start with the proposition you've got a pool of educated individuals working in STEM-type jobs, and go from there. Then you at least know you're working with a pool of people who might have some chance of being interested in your collection of Star Wars figurines, or who want to debate the relative merits of Jar Jar as a character.
Because, really, if you tell the person you're on a date with that you used Python scripts to categorize people into several containers ... you're not gonna get a second date, and the one you're on might end abruptly as the awkward silence turns into thoughts that you might, in fact, be some kind of creepy stalker.
BEGIN NERD VOICE
I've done stochastic analyses of your responses to questionnaires and exhaustively compared your responses to other women on this site, and I calculate there is an 45.2% you might like me. You're the highest score yet!
END NERD VOICE
Really, don't be that guy.
Lost at C:>. Found at C.
Even with the women in question also choosing him on the basis of his tailored responses, he's simply increasing the sample size (i.e. the number of first dates) he gets, without really addressing the quality of the data - how closely the women match him in reality and vice-versa.
One of his descriptions in the article "star signs and all that crap" (or words to that effect) indicates that he still hasn't really "got" the women in the database. By dismissing what they consider important in a profile (the "crap") he's not helping himself. Maybe he should have turned around his search. Instead of hacking his profile to get more matches, he should consider modifying his personality to be more attractive to what the larger numbers of women feel they want in a man.
But I guess to a techie, every problem has a technical solution. No doubt all the first-date restaurants will thank him for his patronage and his (later, but maybe not much later) divorce lawyer will also be suitably grateful.
politicians are like babies' nappies: they should both be changed regularly and for the same reasons
So far no results for me. They all seem to want "rich, handsome guy who loves to travel."
The cow says "Moo." The dog says "Woof." The Timothy says "Thanks, valued customer. We appreciate your input."
Even though Howard wanted to say "mumbo jumbo"
Just the headline reminded me of a story I read a couple of years ago. That site needs registration but he also posted it on a site which does not. 99 First Dates. Hilarious.
Warning - Adult Content if you worry about that kind of thing.
Mielipiteet omiani - Opinions personal, facts suspect.
TED is now basically full of pseudoscientific bullshit and ego-fueled self-promoters.
Finding true love is really nothing more than matching your personal with either the same or a complimentary crazy in someone else. Wearing your crazy on the first date is risky, but reduces wasted time on those who are incompatible with your personal brand of insanity.
Is it just my observation, or are there way too many stupid people in the world?
1) Categorize women into 7 categories. That looks to me to be valuable, but the article did not discuss all 7 categories. It ignored the only interesting thing this guy did!
2) Set up multiple profiles and use machines to initiate action with thousands of potential women.
3) Went on hundreds of dates in a relatively short amount of time.
His 'success' was statistically insignificant AND totally unrelated to his math. Anyone that goes on hundreds of dates and find the right woman.
You want to impress me? Have the algorithm pick 5 women and have them all be very interested in you. Picking 100's of women with lots of failed dates is just a NORMAL DATING LIFE.
excitingthingstodo.blogspot.com
Whilst what he did was very clever, at the end of the day he manipulated the scoring so that his profile was placed in front of thousands of womens search results because it had a high match percentage (that normally would never have been seen).
The TL;DR version of this story is that if thousands of women see your profile and, at the same time, are told by a website that you're a high match to them, then you've got a very good chance they'll contact you. Which shouldn't be a surprise to anyone.
That, to me, is the digital equivalent of (the old advice) that you'll never meet someone unless you get yourself out there.
Avantslash - View Slashdot cleanly on your mobile phone.
now?
They always start out with an super-long, totally unreasonable requirements list that includes stuff that's totally irrelevant to any imaginable scope. Through hard work and negotiation and development of what you initially bring to the table, you need to bring down the client's impossible functional specification to something workable she can reasonably be satisfied with (also beer helps). It's called "game" for a reason.
Medical innovation comes from, get this, medical research. I'm not going to buy the fundamental credibility of someone who hasn't even necessarily seen the inside of a university, much less a medical school. If you see a "cold medicine invented by a teacher" altmed package at the checkout of a grocery store, you don't go "oh anyone could come up with something that works" you flip it over and see "*this statement not evaluated by the FDA" because it's bullshit.
Yeah, but under the lens of actual medical science it all falls apart.
From wikipedia
Many of Andraka's claims do not stand up to rigorous peer-reviewed research. For instance, a 2011 article published by Sharton et al. of the Laboratory of Molecular Biology at the NIH National Cancer Institute refutes many of Andraka's claims about specificity of using mesothelin as a biomarker for pancreatic cancer. Specifically, the group showed that mesothelin serum levels in healty donors 0.58 (0.15 – 0.72) nmol/l were not statistically different from serum levels in pancreatic cancer patients 0.66 (0.52 – 0.94) nmol/L.[15] In addition to this issue of false positives, George M. Church, professor of genetics at Harvard University, has raised concerns about the cost, speed, and sensitivity claims.[11]
However I agree with his approach to keep it short and simple.
Fuck....now you tell me. This date has lasted 8 years, 3 cars, 2 houses, and 3 kids. She just won't take a hint....but I don't want to be rude.
You can discuss exceptions, but the reality is that red flags on pseudoscience are important for casual observations of this sort.