Porbability, not sure I get it

  • Thread starter Thread starter liquidFuzz
  • Start date Start date
AI Thread Summary
The discussion revolves around calculating probabilities in string matching using Bayes' theorem. The user is trying to determine the probability of a true hit (A) versus the probability of an artifact being correct (B) but finds the calculations confusing. There is a request for clearer examples and a general description of the probabilities involved, as the current context lacks clarity. Participants express concerns about the validity of probability values exceeding one and the need for more concrete definitions of the scenarios being analyzed. Overall, the conversation highlights the importance of precise definitions and examples in probability calculations.
liquidFuzz
Messages
107
Reaction score
6
An example of what I try to calculate. In a string matching between two strings the whole strings aren't compared, just a section.

A = The probability of a true hit.
A true hit being:
PHP:
|adcbacdcbacd|abacd
|adcbacdcbacd|ab

B = The probability of artifact being correct.
Falls artifact being:
PHP:
|adcbacdcbacd|abacd
|adcbacdcbacd|bca

I though I'd use Bayes theorem to calculate the probability, but I'm not sure of how to do it.
P(A|B)\frac{P(B|A)P(A)}{P(B)}

In my example A is much higher than B, would this imply:
P(A|B)\frac{P(≈A)P(A)}{P(B)} ≈ P(A)
 
Mathematics news on Phys.org
I don't get what you want to calculate.
Can you add a general description instead of 2 uncommented examples?
 
mfb said:
I don't get what you want to calculate.
Can you add a general description instead of 2 uncommented examples?
A = Probability of something happening
B = Probability of something goes wrong if A happened.
 
That is not the problem.
What do you want to calculate, where do those probabilities come from, and so on?

B = Probability of something goes wrong if A happened.
So P(B) is P(goes wrong|A)?
 
Ok, I snitched the text that got my puzzled. Please comment!
 

Attachments

  • Bayes.png
    Bayes.png
    30.7 KB · Views: 461
Did you have a particular question in mind given your snitch?
 
Yeah, I don't quite get the part when he or she goes all P(A|B) and P(B|A).

This is a part of a report that I'm helping to proofread.
 
Probability values above 1 do not make sense. In addition, the text does not state clearly where probabilities come from. It is like "probability of an apple" - sure, I know apples, but an apple where, in which setup?
 
I guess I'll get away with just commenting this. "Vague reasoning."

Thanks!
 
Back
Top