Overfitting can be seen in machine learning when a statistical model describes random error or noise instead of the underlying relationship. Overfitting is usually observed when a model is excessively complex. It happens because of having too many parameters concerning the number of training data types. The model displays poor performance, which has been overfitted.
