The Confusion Matrix

Welcome to **Day 147**. Accurate models can still be dangerous.

The Scenarios

1. **False Positive**: The doctor says you're pregnant, but شما're a man. (Annoying, but not fatal).

2. **False Negative**: The security alarm doesn't go off during a robbery. (**Catastrophic**).

It's a 2x2 grid that shows:

**True Positives**: Correctly predicted "Yes."

**True Negatives**: Correctly predicted "No."

**False Positives**: Incorrectly predicted "Yes."

**False Negatives**: Incorrectly predicted "No."

from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test, predictions)

sns.heatmap(cm, annot=True)

**Precision**: When I say it's Spam, how often am I right?

**Recall**: Of all the Spams out there, how many did I actually catch?

Generate a confusion matrix for your classification model and identify how many "False Positives" it produced.

*Day 148: K-Means Clustering (Unsupervised Learning).*