What does overfitting in machine learning models mean?

Correct answer: The model performs well on training data but poorly on unseen test data because it has memorized patterns specific to training data

Option C is correct because overfitting occurs when a model learns the training data too specifically, capturing noise and data-specific patterns rather than generalizable relationships, which results in high accuracy on training data but poor performance on unseen test or production data. Option A, requiring too much computational power, describes a resource or scalability concern unrelated to the statistical concept of overfitting. Option B, failing to learn any patterns from training data, describes underfitting, the opposite problem where the model is too simple to capture the underlying signal. Option D, using too many features, can contribute to overfitting but is not the definition of overfitting itself; the definition centers on the generalization gap between training and test performance.