Explanation:
In Spark ML, Class Weighting is a technique designed to tackle class imbalance by assigning different weights to classes, thereby emphasizing the minority class during model training. This approach enhances the model's sensitivity towards the under-represented class without resorting to more complex methods like over-sampling, under-sampling, or SMOTE. It's a straightforward yet powerful method to improve classification accuracy on imbalanced datasets.
Ultimate access to all questions.
No comments yet.
When working on a machine learning project with imbalanced classes, which Spark ML supported technique can be employed to ensure accurate model training?
A
Synthetic Minority Over-sampling Technique (SMOTE)
B
Under-sampling
C
Class Weighting
D
Over-sampling