Understanding True Positives, False Positives, True Negatives & False Negatives

This sounds silly now and I am almost ashamed to say how much of a terrible time I have always had remembering what constitutes true positives, true negatives, false positives & false negatives.

To this day - I will slip occassionally but much lesser so now then before!

Here is the silly way it finally worked and clicked for me:

True positives == these are actually true AND have been selected by the model. Yay, found a good one!

True negative == these are actually wrong and have been classified as wrong by the model. Nothing here, really!

False Negatives == these are actually correct but have been classified as wrong by the model. Missed a good one, so sorry!

False Positive == these are actually wrong but have been classified as correct by the model. Oops, false alarm, sorry!.


This diagram from Wikimedia also really helped click & remember this concept. I am a visual learner, as such the picture is often stuck in my head.

Notes mentioning this note

There are no notes linking to this note.