Asking a girl on a second date.

ayz said:
my roommate was a math major and took that class, they did a bunch of obscure abstract proofs like "prove numbers exist" or some stupid shit, the kid would do his homework and have 5-6 pages for a single proof.
Click to expand...

:lol: I'm in "that class" right now... :(
 
SegaRob said:
These FP and FN are both WRONG answers, but depending on the relative number of FP to FN, the answer to raven's question is different. The parameter raven asked for is known as the true positive rate, TPR, which is defined as the number of true positives out of the total number of positives identified by the test.

Mathematically that can be written as:

TPR = FP / (TP + FP)

However, FP is not uniquely defined for a given value of acc. A second parameter that is independent of acc must be defined in order to determine FP and hence TPR. Based on the value of FP, TPR can range from 0 to 1, the full domain for TPR.

And I wasn't even a math major.
Click to expand...

oh, i get it; and like i said before, unless its explicitly stated, i've always treated the accuracy such that the FP is essentially 0. So for the purposes of raven's posed question, i have been treating the "false correct answer" probability as 0.

in the real world, if you have a device that is 95% accurate, theres no way to measure its FP. this device gives the right answer 95% of the time and the wrong answer 5% of the time. maybe its only 90% accurate and 5% of the time it is giving the right answer even though its a 'false right answer'. but theres no way to determine that. so if you can guarantee it gives the right answer 95% of the time, regardless of how much of that percentage is a 'FP', its 95% accurate.

What you are saying is correct and totally makes sense, and you can nitpick on the theoretical side all you want, but this sounds like an argument you would only deal with in a textbook.
 
Eggi said:
oh, i get it; and like i said before, unless its explicitly stated, i've always treated the accuracy such that the FP is essentially 0. So for the purposes of raven's posed question, i have been treating the "false correct answer" probability as 0.

in the real world, if you have a device that is 95% accurate, theres no way to measure its FP. this device gives the right answer 95% of the time and the wrong answer 5% of the time. maybe its only 90% accurate and 5% of the time it is giving the right answer even though its a 'false right answer'. but theres no way to determine that. so if you can guarantee it gives the right answer 95% of the time, regardless of how much of that percentage is a 'FP', its 95% accurate.

What you are saying is correct and totally makes sense, and you can nitpick on the theoretical side all you want, but this sounds like an argument you would only deal with in a textbook.
Click to expand...
this isn't true - FP is something you can measure. for instance, in the real world, you can do mammography that gives a positive test for breast cancer. then you do a biopsy and realize it's not breast cancer. that is a false positive.

in general, you need a "gold standard" test that is 100% accurate to know the TRUE state. you don't use it because it's expensive, inconvenient or whatever so you have cheaper, more convenient tests with less accuracy that you use for screening.

if FP=0, then the answer to raven's question would be 100%.
 
Last edited:
SegaRob said:
this isn't true - FP is something you can measure. for instance, in the real world, you can do mammography that gives a positive test for breast cancer. then you do a biopsy and realize it's not breast cancer. that is a false positive.

in general, you need a "gold standard" test that is 100% accurate to know the TRUE state. you don't use it because it's expensive, inconvenient or whatever so you have cheaper, more convenient tests with less accuracy that you use for screening.

if FP=0, then the answer to raven's question would be 100%.
Click to expand...

ok, so i was misinterpreting what you were trying to say and glossed over your explanation on the last page...

i thought you were implying that there were essentially 6 possible outcomes
ex: machine tests for mormons, the subject is mormon, the machine returns a positive answer (correctly), however it returned this positive answer as part of the inaccuracy - ie it didnt correctly detect the person as mormon, or something... i was thinking that was what you meant by FP, as in a false positive for detecting a mormon person mormon...which made no sense to me; so i'm glad i just completely misinterpreted you.

SegaRob said:
Code: 
              |         Mormon      |       Not Mormon
------------------------------------------------------
Test Positive |    [b]True Positive[/b]    |    [b]False Positive[/b]
------------------------------------------------------
Test Negative |   [b]False Negative[/b]    |    [b]True Negative[/b]
Click to expand...

so back to this; if FN and FP are the same (which i would always assume if they arent explicitly stated in the problem) then we are on the same page.

good.

SegaRob said:
Total False = FP + FN = 5% * population

a) this means you can't know FP without knowing more information.
b) FP and FN are interrelated by the equation above, so you cannot just look at FP.

depending on the ratio of FP/FN, the answer to the question could range from 0% to 100%.
Click to expand...

a) should be assumed if only the accuracy is given flatly as "detects accurately 95% of the time"; therefore 5% rate for FP, and 5% rate for FN;
b) yea like i said id always assume they were the same unless explicitly stated in the problem.

so its a relief to me that i was just confused by your argument. we have essentially agreed for the last 3 pages, but you've just been arguing the problem wasnt defined clearly enough for you :)
 
Last edited:
SegaRob said:
this isn't true - FP is something you can measure. for instance, in the real world, you can do mammography that gives a positive test for breast cancer. then you do a biopsy and realize it's not breast cancer. that is a false positive.

in general, you need a "gold standard" test that is 100% accurate to know the TRUE state. you don't use it because it's expensive, inconvenient or whatever so you have cheaper, more convenient tests with less accuracy that you use for screening.

if FP=0, then the answer to raven's question would be 100%.
Click to expand...

If FP = 0, then the overall accuracy of the test could not possibly be 95%. A test that returned negative on every single person would have FP = 0 and an overall accuracy of 98.1% since only 1.9% of Americans are mormon. Are you really this fucking dumb?

I'm wondering if you were really too dumb to interpret the question or if you are just a bad troll
 
-Bane said:
:lol: I'm in "that class" right now... :(
Click to expand...

yeah i had bme and econ as majors, i didnt feel the need to pick up a math major by taking that shitty class

all my roommate would do is bitch about how conceptually difficult yet completely fucking pointless his homework was
 
-raven- said:
If FP = 0, then the overall accuracy of the test could not possibly be 95%. A test that returned negative on every single person would have FP = 0 and an overall accuracy of 98.1% since only 1.9% of Americans are mormon. Are you really this fucking dumb?

I'm wondering if you were really too dumb to interpret the question or if you are just a bad troll
Click to expand...

If you had FN=15,000,000 & FP=0, acc=95%.

If you're just going to insult me when I'm clearly right, I'm going to just put you on ignore.
 
Eggi said:
a) should be assumed if only the accuracy is given flatly as "detects accurately 95% of the time"; therefore 5% rate for FP, and 5% rate for FN;
b) yea like i said id always assume they were the same unless explicitly stated in the problem.
Click to expand...

There is absolutely no reason to assume the test is equally good at detecting Mormons as it is detecting not Mormons. In general, it is NOT true for a test. Moreover, we cannot even consider it as a conservative estimate since without making the assumption the answer can take on all possible values in the domain.
 
SegaRob said:
There is absolutely no reason to assume the test is equally good at detecting Mormons as it is detecting not Mormons. In general, it is NOT true for a test. Moreover, we cannot even consider it as a conservative estimate since without making the assumption the answer can take on all possible values in the domain.
Click to expand...

lol are you daft?

he made the problem up, theres no such thing as a mormon detector. its a hypothetical.

And there is absolutely no reason to assume it isnt equally good at detecting FP and FN seeing as he didnt explicitly state otherwise. i am glad you found something to argue about with ravens problem which defends you telling him he is incorrect, but now this is getting stupid.

Do you/did you sit there in class and nitpick with a professor when they made an approximation? Do you/did you sit there during a test and not answer a test question because every assumption wasnt explicitly written out? jesus, you wouldnt survive a day in an engineering class.

ps: i am done here. SoD you should update this thread after your second date when you start bleeding from the mouth and she tells you to have a nice life.

pps: if the last 6 pages were deleted, this thread would still be humorous.
 
Last edited:
Eggi said:
lol are you daft?

he made the problem up, theres no such thing as a mormon detector. its a hypothetical.

And there is absolutely no reason to assume it isnt equally good at detecting FP and FN seeing as he didnt explicitly state otherwise. i am glad you found something to argue about with ravens problem which defends you telling him he is incorrect, but now this is getting stupid.

Do you/did you sit there in class and nitpick with a professor when they made an approximation? Do you/did you sit there during a test and not answer a test question because every assumption wasnt explicitly written out? jesus, you wouldnt survive a day in an engineering class.

ps: i am done here. SoD you should update this thread after your second date when you start bleeding from the mouth and she tells you to have a nice life.
Click to expand...
This isn't anything like an approximation. An approximation would be saying, well a supercomputer can do this numerically and come up with 0.0009, but we can do it with simple algebra and get 0.001. This is saying the answer is ANYTHING, but for argument's sake we'll say it 25%. You don't seem to realize that this assumption is not reasonable AT ALL.

I wouldn't survive a day on anything you engineered if you made arbitrary assumptions like this.
 
Eggi said:
lol are you daft?

he made the problem up, theres no such thing as a mormon detector. its a hypothetical.

And there is absolutely no reason to assume it isnt equally good at detecting FP and FN seeing as he didnt explicitly state otherwise. i am glad you found something to argue about with ravens problem which defends you telling him he is incorrect, but now this is getting stupid.

Do you/did you sit there in class and nitpick with a professor when they made an approximation? Do you/did you sit there during a test and not answer a test question because every assumption wasnt explicitly written out? jesus, you wouldnt survive a day in an engineering class.

ps: i am done here. SoD you should update this thread after your second date when you start bleeding from the mouth and she tells you to have a nice life.

pps: if the last 6 pages were deleted, this thread would still be humorous.
Click to expand...

So, you're saying that when someone poorly defines a math question it's the people who he asks to answer it's problem that its poorly defined?

retarded

Theres a difference between approximating .49 as .5 and assuming that two things that are virtually never equal are equal.
 
SegaRob said:
This isn't anything like an approximation. An approximation would be saying, well a supercomputer can do this numerically and come up with 0.0009, but we can do it with simple algebra and get 0.001. This is saying the answer is ANYTHING, but for argument's sake we'll say it 25%. You don't seem to realize that this assumption is not reasonable AT ALL.

I wouldn't survive a day on anything you engineered if you made arbitrary assumptions like this.
Click to expand...

lol my point is that you're being a nitpicky douche, but thanks for telling me what an approximation is

Darkstrand said:
Theres a difference between approximating .49 as .5 and assuming that two things that are virtually never equal are equal.
Click to expand...
what? i only mentioned the approximation thing because i picture segarob as one of those nitpicky douchebags back in college that would ask tons of questions when a problem didnt explicitly state every assumption

"well sure i know you are supposed to open the peanut butter jar with your hand, but it doesnt say that in the directions!!!"

anyways, i consider an assumption like the FP/FN being equal unless explicitly stated otherwise to be a realistic assumption to make seeing as thats what i used throughout college :shrug: w/e no matter.
 
Last edited:
You must log in or register to reply here.
Back
Top