Overfitting is a modeling error where a machine learning model performs well on training data but poorly on new data. Used in ML to identify and correct excessively complex models.
Example:
A model trained on a small dataset predicts perfectly for training but fails on test data.
