Scale and Normalize: When to Use Each Method?

Quick Overview

This interview question evaluates core ML concepts, assumptions, math intuition, training/evaluation trade-offs, and practical failure modes in a realistic interview setting. A strong answer for Scale and Normalize: When to Use Each Method? states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

Scale and Normalize: When to Use Each Method?

Company: Boston Consulting Group

Role: Data Scientist

Category: Machine Learning

Difficulty: easy

Interview Round: Take-home Project

##### Scenario BCG CodeSignal notebook – feature scaling step before modeling ##### Question Given a DataFrame df with numeric columns age and income, demonstrate how to Standard-scale age and Min-Max normalize income. Explain when you would prefer each scaler. ##### Hints Use sklearn.preprocessing; relate to Gaussian vs bounded distributions.

Quick Answer: This interview question evaluates core ML concepts, assumptions, math intuition, training/evaluation trade-offs, and practical failure modes in a realistic interview setting. A strong answer for Scale and Normalize: When to Use Each Method? states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

|Home/Machine Learning/Boston Consulting Group
Boston Consulting Group logo
Boston Consulting Group
Aug 4, 2025, 10:55 AM
easyData ScientistTake-home ProjectMachine Learning
5
0

Scale and Normalize: When to Use Each Method?

Feature Scaling Before Modeling (CodeSignal Notebook)

Context

You're preparing features in a notebook step before training a model. You have a pandas DataFrame df with two numeric columns: age and income.

Task

  1. Standard-scale the age column (mean 0, variance 1).
  2. Min-Max normalize the income column to [0, 1].
  3. Briefly explain when you would prefer each scaler.

Assume scikit-learn is available.

Clarifying Questions to Ask Guidance

  • Clarify the task, data shape, labels, constraints, and evaluation metric.
  • State assumptions behind the math or modeling technique you choose.
  • Connect theory to practical training, debugging, and deployment implications.

What a Strong Answer Covers Guidance

  • Correct definitions and formulas where the prompt requires them.
  • A practical explanation of how the method behaves on real data.
  • Trade-offs, failure modes, diagnostics, and mitigation strategies.
  • Evaluation choices that match the product or modeling objective.

Follow-up Questions Guidance

  • How would noisy labels, class imbalance, or distribution shift affect the answer?
  • What would you monitor after deployment?
  • Which baseline would you compare against first?
Loading comments...

Browse More Questions

More Machine Learning•More Boston Consulting Group•More Data Scientist•Boston Consulting Group Data Scientist•Boston Consulting Group Machine Learning•Data Scientist Machine Learning