Backward elimination

With backward elimination, we start with all the variables. Iteratively, we need to test the elimination of each of the variables. The variable, once again, is deleted with the predefined threshold or criteria. The variables that have the least significant impact on the model's accuracy are eliminated one by one in this method.

It is also possible to utilize both methods together for faster parameter tuning. 

Get Artificial Intelligence for Big Data now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.