This is the 9th day of my participation in the August More Text Challenge
This paper introduces a series of basic evaluation indicators of machine learning.
Basic definition
- T: True indicates that the judgment is correct
- F: False indicates incorrect judgment
- P: PostIve indicates that the sample is judged to be positive
- N: Negative indicates that the sample is Negative
Index definition
If you are always confused, translate the meaning according to the alphabetical order above.
- TP: (T) The judgment is correct, (P) the sample is judged to be positive (in fact, the sample is positive)
- TN: (T) The judgment is correct, and (N) the sample is judged to be negative (in fact, the sample is negative)
- FP: (F) The judgment is wrong, (P) the sample is judged to be positive (in fact, the sample is negative)
- FN: (F) The judgment is wrong, (N) the sample is judged to be a negative sample (in fact, the sample is positive)
| Evaluation indicators | Predicted results | ||
|---|---|---|---|
| Is the sample | Negative samples | ||
| The actual situation |
Is the sample | TP | FN |
| Negative samples | FP | TN | |
Deepen the understanding
-
TP and TN are the cases judged correctly by the discriminator, and the original positive/negative samples are classified correctly
-
FP means that the negative sample is mistaken for the positive sample, which is a false alarm
-
FN indicates that positive sample is mistaken as negative sample, indicating alarm leakage