Is it good to label the data based on sub category than parent category? For example: for drugs data ... label the drugs dose as drug_dose or label the drug dose as different type of dose like injection, capsule, spray..which is a better option?
While labeling, label with sub-categories. This enables you to work with larger set of tools / algos.
- With Algos that work better with linear labels (and don't know about label hierarchies), you can pre-process to transform and use only least granular labels (E.g.: CoreNLP)
- With some of Deep Learning models, you can use sub category tags as it is.
Since tagging / labeling is expensive and (potentially one-time) exercise; it is better to tag more granular categories.