When do you add interaction terms versus switching to a nonlinear model?
PICTURE THIS: DOM IS A TREE
Simple meaning
Add a few domain interactions when you need a still-interpretable linear log-odds model.
WHY — Logistic Regression instead of guessing?
Why interviewers care about Logistic Regression:
contrast on Logistic Regression, not two memorised paragraphs.
the developer, then one case where picking wrong hurts.
Name the idea, why it exists, then one short example.
End with when you use it and one common pitfall.
STEPS — What happens step by step?
Before you speak the answer, walk the interviewer through these steps:
- 1Add a few domain
interactions when you need a still-interpretable linear log-odds model.
- 2If you need many
unknown interactions, trees or a small net will discover them with less manual search.
- 3Always compare a well-regularized
logistic baseline before celebrating the nonlinear model.
- 4Give an example
One tiny concrete case you can say aloud.
- 5Common mistake
What juniors usually get wrong.
- 6Close
When you pick this over the alternative.
EXAMPLE — See it in action
Here's a short line you can speak, broken into clear beats:
Note: Adapt this scaffold to your own project — keep it under 60–90 seconds.
Key takeaway
Add a few domain interactions when you need a still-interpretable linear log-odds model. If you need many unknown interactions, trees or a small net will discover them with less manual search.