I have data set which contains 14 categorical nominal data. However, the dataset is imbalanced some categories have more sample than others. Data were one-hot encoded and I tried SMOTE-NC. I'm not sure if this technique is relevant to categorical data only because is K-nearest based.
Could someone suggest me a technique or solution for categorical nominal dataset only?
Related to your query, some beautiful incitements in this tutorial: https://www.datacamp.com/community/tutorials/categorical-data may help you.
See also https://data.world/datasets/categorical
Thank you @Muhammad Ali, but those tutorials are not related to my problem.@
https://towardsdatascience.com/handling-imbalanced-datasets-in-deep-learning-f48407a0e758
https://www.analyticsvidhya.com/blog/2017/03/imbalanced-classification-problem/
https://machinelearningmastery.com/tactics-to-combat-imbalanced-classes-in-your-machine-learning-dataset/
I have similar problem and I have applied the SMOTE-NC and it works perfectly after encoding the categorical variables but I don't know if that true. Did you reach any solution so far regarding this problem?? Tobiasz Kaduk
- Badges
- Science topic