Your question is Feature Engineering for NLP Models. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
You are building an NLP classifier and need to decide which text features to use before training the model. The input text is noisy, short, and inconsistent, so preprocessing choices can change the result a lot.
What are the best practices for feature engineering in natural language processing?