suppose you have unbalance image dataset.you decided to increase the sample size of minority class by duplicating samples,will the classifier solve bias variance problem if you use Newly created dataset that contains duplicate value? if yes ,why? if no,why not?
Duplication will only compound any distributional error.
Better to ensure that the 'minority' is proportionally representative of population.
Hope this helps,
Matt
Matthew Kerry thanks for your comment,is it possible to implement something like smote in image classification task? if yes,can you please refer to any python code example ?
Dear Syed Mobassir Hossen
It’s a very common problem in machine learning and deep lear algorithms. Please read the following:
https://machinelearningmastery.com/tactics-to-combat-imbalanced-classes-in-your-machine-learning-dataset/
Hope it helps,
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size