Over-fitting of Machine Learning model

[Abhiram.m]

When a model performs well on training data but poorly on untried data, this is over-fitting. 

Over-fitting is a very common problem in Machine Learning. There is an extensive range of literature dedicated to studying methods for preventing over-fitting.
Three simple steps to avoid Over-fitting:

• Train with more data - Training with more data can help algorithms detect the signal better (But, it won’t work every time).
  
• Keep the model simple - take fewer variables into account, thereby removing some of the noise in the training data
  
• Utilize cross-validation methods, such as the k-folds method.
  
• Use regularization techniques - Regularization describes a wide range of methods for forcibly simplifying your model, such as LASSO, that penalize certain model parameters if they're likely to cause over-fitting.
  
• Ensembling - Machine learning techniques called ensembles are used to combine predictions from various independent models.