Return to Video

Confusion Matrix for Eigenfaces - Intro to Machine Learning

  • 0:00 - 0:03
    So talk about confusion matrices in different context, you might
  • 0:03 - 0:08
    remember the work that Katie showed you on principle component analysis, on PCA.
  • 0:08 - 0:13
    Were we looked at seven different, I guess white male politicians from
  • 0:13 - 0:17
    George Bush to Gerhard Schroeder, and we ran an eigenface analysis.
  • 0:17 - 0:22
    Extracting the principle components of this data set, and then re-use
  • 0:22 - 0:28
    the eigenfaces to re-map new faces to names in order to identify people.
  • 0:28 - 0:31
    So what I'm going to do now, I won't drag you
  • 0:31 - 0:34
    through the same PCA example again, so let's do away with those faces.
  • 0:34 - 0:38
    But instead what I'll do is, I give you a typical output and
  • 0:38 - 0:42
    we're going to study the output using confusion matrices.
  • 0:42 - 0:49
    So Katie was so nice to run a PCA on the faces of those politicians and
  • 0:49 - 0:53
    take the resulting features, put it into a support vector machine and
  • 0:53 - 0:54
    then go through the data and
  • 0:54 - 1:00
    count how often any of those people were predicted correctly or misclassified.
  • 1:00 - 1:03
    And just to confuse everybody, in this example,
  • 1:03 - 1:06
    we follow the convention of putting the true names, the true class labels,
  • 1:06 - 1:08
    on the left, and the predicted ones on top.
  • 1:08 - 1:13
    So, for example, this number one over here was truly Donald Rumsfeld, but
  • 1:13 - 1:15
    was mistaken to be Colin Powell.
  • 1:15 - 1:19
    And the way I know it's Colin Powell was the same names over here,
  • 1:19 - 1:23
    from Ariel Sharon to Tony Blair, apply to the columns over here.
  • 1:23 - 1:26
    Ariel Sharon's on the left and Tony Blair's on the right.
  • 1:26 - 1:28
    So we ask you a few questions now.
  • 1:28 - 1:29
    First, a simple one.
  • 1:29 - 1:33
    Which of those seven politicians was most frequent in our dataset?
Title:
Confusion Matrix for Eigenfaces - Intro to Machine Learning
Description:

more » « less
Video Language:
English
Team:
Udacity
Project:
ud120 - Intro to Machine Learning
Duration:
01:34

English subtitles

Revisions Compare revisions