How would you handle an imbalanced dataset?
Question
Explain Tactics to get over the humps
in progress
0
Machine Learning
55 years
1 Answer
741 views
Contributor 0
Answer ( 1 )
There are several techniques to handle imbalanced dataset:
We can use random undersampling where the number of instance of majority class is deleted.
We can also use random oversampling where the number of instance of minority classis duplicated
We can use SMOTE ( Synthetic Minority Oversampling Technique) to add instance of minority class. SMOTE uses nearest neighbours of the minority class to create synthetic data.