0:00
/
0:00
Transcript
0:00
SPEAKER 1
Confusion matrix can be confusing. More specifically, how do we interpret the four entries of a confusion matrix, which are true positive, false positive, false negative, and true negative? Consider a cat-dog classifier. The cat being a positive class here, if for a particular sample, the true label is cat, but the model predicted it as dog,
0:22
what type of prediction is this? Is this true positive, false positive, false negative, or true negative? Are you confused? If yes, let me share a cool technique that will help you answer this real quick. Here's what you have to do. When labeling any binary classification prediction, you just need to ask these two questions.
0:40
The first question you need to ask is, did the model get it right? Quite intuitively, the answer will be either yes or no. Yes means true and no means false. The next question you need to ask is, what was the predicted class? Here, the answer will be either positive or negative.

A Technique to Understand TP, TN, FP and FN

An intuitive guide to label predictions.

Do you struggle to classify predictions as one of TP, TN, FP, and FN?

If yes, here’s a simple technique I recommend using.

A video version of this technique is available at the top.

If you prefer reading...let’s dive in below.


When labeling any binary classification prediction, ask two questions:

  • Question 1) Did the model get it right?

    • The answer will be either Yes or No.

      • Yes means True.

      • No means False.

  • Question 2) What was the predicted class?

    • The answer will be either Positive or Negative.

Next, just combine the above two answers to get the final label.

For instance, say the actual and predicted class were positive.

  • Question 1) Did the model get it right?

    • Answer: Yes, which means TRUE.

  • Question 2) What was the predicted class?

    • Answer: POSITIVE.

The final label: TRUE POSITIVE.

Simple, right?

The following visual summarizes this:

As an exercise, complete the table below.

Consider:

  • The cat class → Positive.

  • The dog class → Negative.

Let me know your answers.

👉 Over to you: Do you know any other techniques to label binary classification predictions?

Thanks for reading!


P.S. For those wanting to develop “Industry ML” expertise:

At the end of the day, all businesses care about impact. That’s it!

  • Can you reduce costs?

  • Drive revenue?

  • Can you scale ML models?

  • Predict trends before they happen?

We have discussed several other topics (with implementations) in the past that align with such topics.

Develop "Industry ML" Skills

Here are some of them:

All these resources will help you cultivate key skills that businesses and companies care about the most.