I have been doing some reading on data pre-processing and feature engineering including feature selection, feature importance and feature construction. My understanding is that Feature engineer is applied in data preprocessing stage. Additionally, Feature importance is sometimes checked when applying machine learning models, some times ot is done internally in the modes.
My question: Is feature engineering always implemented in pre-processing stage or sometimes could be performed in later stages?
Thank you Shosho
Feature engineering is the process of creating new features using domain knowledge, where the term 'pre-processing is meant to data cleaning.
Here, answering your question there is no specific order for feature engineering process. it depends on the data and domain. you have to keep checking after creating new features and retrain the model with new features. Normally performing feature engineering in early stages is preferred.