Confusion Matrix

Tags

artificial-intelligence

Is a way to visually represent the accuracy of a classifier based on sampled training data.

Along the horizontal axis we place the actual class of each sample, along the vertical axis we place what our classifier output for that sample.

For example:

     CAT DOG
CAT | 5 | 2 |
DOG | 3 | 3 |

From our definition above and this example we can tell that the classifier was given 8 pictures of cats and 5 of dogs. Of those it misclassified 3 cats as dogs and 2 dogs as cats.

We can also tell that out of the 13 samples given, the classifier got \(5+3=8\) correct.