Which issue arises when the model fits the training data extremely well but does not generalize to new data?

• A. Underfitting
• B. Overfitting
• C. Recall
• D. F-score

Answer: (B)

Overfitting happens when a model learns the training data too well, including noise and random fluctuations, so it performs exceptionally on that data but fails to generalize to new, unseen data. This usually occurs when the model is too complex relative to the amount of training data, causing it to memorize specifics rather than capture the underlying pattern. The result is high accuracy on the training set but poor performance on new data, indicating high variance.

To improve generalization, you can simplify the model, gather more data, or apply regularization and validation techniques such as early stopping, dropout, or pruning. The other terms don’t describe this problem: underfitting means the model is too simple to capture the patterns, leading to poor performance even on training data; recall and F-score are evaluation metrics, not issues of how well a model generalizes.