Translation: Overfitting occurs when a model learns the training data too closely, including noise or random fluctuations that do not represent general patterns. As a result, the model performs well on the training data but fails to generalise, leading to poor performance on new, unseen data.