Knowledge MaxxingClass imbalance
Class imbalance occurs when one class in a classification dataset is far more common than another, which can bias a model toward always predicting the majority class.
Prerequisite: Accuracy
What matters
Standard accuracy becomes misleading under imbalance, since a model can score high just by favoring the majority class, so other metrics and techniques are needed to properly judge and address it.
Example
In a fraud dataset where only 1% of transactions are fraudulent, a model that always predicts 'not fraud' looks 99% accurate while catching zero fraud.
Common misconception
Class imbalance does not always need to be 'fixed'; the right approach depends on whether the minority class actually matters more for the task's real costs.
Recall question
Why does high accuracy become misleading on an imbalanced dataset?