Return to Video

Bayes Rule Diagram - Intro to Machine Learning

  • 0:00 - 0:05
    What we really said that we had a situation that prior
  • 0:05 - 0:10
    a test is a certain sensitivity and a certain specificity.
  • 0:10 - 0:15
    When you receive say a positive test result, what you do is you take your prior,
  • 0:15 - 0:19
    you multiply in the probability of this test result.
  • 0:19 - 0:24
    Given C, and you multiply in the probability of the test result given not C.
  • 0:24 - 0:28
    So this is your branch for the consideration that you have cancer.
  • 0:28 - 0:32
    This is your branch for the consideration you have no cancer.
  • 0:32 - 0:36
    When you're done with this, you arrive at a number that now combines the cancer
  • 0:36 - 0:38
    hypothesis with the test result.
  • 0:38 - 0:41
    Both for the cancer hypothesis and the not cancer hypothesis.
  • 0:43 - 0:46
    Now what you do, you add those up.
  • 0:46 - 0:50
    And they normally don't add up to one.
  • 0:50 - 0:51
    You get a certain quantity,
  • 0:51 - 0:55
    which happens to be the total probability that the test is what it was.
  • 0:55 - 0:57
    This case positive.
  • 0:57 - 0:59
    And all you do next is divide or
  • 0:59 - 1:04
    normalize this thing over here by the sum over here.
  • 1:04 - 1:06
    And the same on the right side.
  • 1:06 - 1:08
    The divider is the same for
  • 1:08 - 1:13
    both cases because this is your cancer range, your non cancer range.
  • 1:13 - 1:15
    But this guy doesn't rely on the cancer variable anymore.
  • 1:15 - 1:20
    What you now get out is the desired posterior probability, and
  • 1:20 - 1:24
    those add up to one if you did everything correct as shown over here.
  • 1:24 - 1:26
    This is your algorithm for Bayes Rule
Title:
Bayes Rule Diagram - Intro to Machine Learning
Description:

more » « less
Video Language:
English
Team:
Udacity
Project:
ud120 - Intro to Machine Learning
Duration:
01:27

English subtitles

Revisions Compare revisions