AI Terms Dictionary

Definition

A confusion matrix is a specific table layout that allows visualization of the performance of an algorithm, typically a supervised learning one. It shows the counts of true positive, true negative, false positive, and false negative predictions. This structure helps in understanding where the model is making errors, providing insights beyond simple accuracy metrics, especially in imbalanced datasets. It serves as the foundation for calculating precision, recall, and F1 scores.

Summary

A table used to describe the performance of a classification model on a set of test data.

Key Concepts

• True Positives
• False Negatives
• Precision
• Recall

Use Cases

• Evaluating binary classifiers
• Analyzing multi-class classification performance
• Debugging model bias in imbalanced datasets

Code Example

1
2
3
4

from sklearn.metrics import confusion_matrix
y_true = [2, 0, 2, 2, 0, 1]
y_pred = [0, 0, 2, 2, 0, 2]
print(confusion_matrix(y_true, y_pred))

Related Terms

• precision
• recall
• F1 score
• ROC curve