Practice English vocabulary for ML feature engineering: normalization, one-hot encoding, feature interactions, feature selection, and feature importance.
0 / 11 completed
1 / 11
What is the difference between 'min-max normalization' and 'z-score normalization'?
Feature normalization prevents features with large scales from dominating distance-based models. Min-max: (x - min) / (max - min). Z-score: (x - mean) / std. Z-score is more robust to outliers since extreme values don't compress the entire distribution into a tiny range.
2 / 11
What is 'one-hot encoding for categorical variables'?
Assigning integers to categories (Red=1, Blue=2, Green=3) implies ordering (Green > Blue > Red) that doesn't exist. One-hot encoding creates separate binary features for each category, letting the model treat them as independent. High-cardinality features (1000+ categories) may use embedding instead.
3 / 11
What is a 'feature interaction' and what signal does it capture?
Some predictive signal only exists in feature combinations. Tree-based models (XGBoost, random forests) automatically learn interactions, but linear models require them to be engineered explicitly. Creating interaction features: multiplication, ratio, or concatenation of two features.
4 / 11
What is 'feature selection' and why does it reduce dimensionality?
High-dimensional feature spaces cause the 'curse of dimensionality' — models see sparse data and overfit. Feature selection methods (filter: correlation/chi-squared; wrapper: recursive feature elimination; embedded: LASSO regularization) remove noise features, improving generalization.
5 / 11
What does 'the feature importance score ranks predictors' mean?
Feature importance (available in Random Forest, XGBoost, etc.) helps practitioners understand the model, identify redundant features for removal, and explain predictions to stakeholders. Common measures: impurity-based importance, permutation importance, and SHAP values.
6 / 11
Code Review Comment: Sarah flagged this PR with the comment: 'Consider adding a feature interaction term between `age` and `income`. The model's performance on younger users seems significantly lower than older ones when income is high. This might indicate a non-linear relationship we're missing.' What does Sarah *most* likely mean in this context? feature interaction
Sarah is highlighting the potential for a non-linear relationship between age and income. Feature interactions allow the model to capture these complex relationships by combining features rather than treating them independently. The correct answer reflects this – creating a new feature (age * income) allows the model to learn that effect.
7 / 11
Liam in the #data-science Slack channel asks: 'I'm building a churn prediction model. The raw data includes 'customer tenure' (days) and 'monthly spending'. I've added these as features. Should I be creating a new feature by multiplying them together? What does this represent?',
This question tests understanding of feature interactions. Multiplying 'customer tenure' by 'monthly spending' creates an interaction term that can capture non-linear relationships – for example, the impact of long-term, high-spending customers might be different from short-term, low-spending ones. It's a common technique to avoid assuming linear combinations are always optimal.
8 / 11
Maria writes in the PR description for her model update: 'I've added an interaction term between `device_type` (mobile/desktop) and `time_of_day`. This accounts for the fact that users on mobile devices tend to use our app less during peak hours.' What is Maria primarily trying to achieve?
Maria's statement highlights a key aspect of feature engineering: capturing non-linear relationships. Interaction terms allow you to represent how the effect of one feature depends on the value of another. While accuracy can be improved, the primary goal here is understanding *why* certain users churn – not just increasing the model's predictive power.
9 / 11
David, during a standup meeting, says: 'I'm creating a new feature called `engagement_score`. It's calculated by multiplying 'number of articles read' by 'time spent reading'. This seems like a good way to capture how engaged users are with our content.' What is David most likely attempting to do?
David is creating an aggregate feature – an 'engagement score' – to represent a complex relationship. This simplification allows for easier understanding and potentially better model performance than using the original features individually. It's important to note that while this approach can be useful, it also risks losing information about the individual components.
10 / 11
Emily is reviewing a PR and sees the following code: '`new_feature = df['age'] * df['income']`.' She notes this creates an interaction term. What potential problem might Emily be observing?
Creating interaction terms can introduce multicollinearity, a situation where two or more features are highly correlated. This makes it difficult to isolate the individual effect of each feature when building and evaluating the model. Multicollinearity leads to unstable coefficient estimates.
11 / 11
Ben reports: 'The feature importance scores for our churn prediction model show that the interaction between 'number of support tickets' and 'customer tenure' is the most important. This suggests that long-term customers with many support tickets are particularly likely to churn.' What does this indicate about the data?
The fact that the interaction term has high feature importance indicates a complex, non-linear relationship. It shows that the model recognizes that the effect of both 'number of support tickets' and 'customer tenure' on churn is not simply additive; rather, their combined impact is significant. This suggests there's an underlying pattern in the data.
What will I practise in "Feature Engineering Vocabulary"?
This module focuses on ML Language — real workplace phrasing you'll use on the job. It contains 11 scenario-based multiple-choice questions with instant feedback.
Is this exercise free to use?
Yes. Every exercise on CoderSlingo, including this one, is free to use with no account or sign-up required.
How many questions does this exercise have?
This module includes 11 questions. Each one gives an immediate right/wrong result plus a full explanation of the correct phrasing.
What happens if I answer a question incorrectly?
You'll see the correct answer highlighted straight away, along with a plain-English explanation of why it's right and why the other options don't fit — mistakes are part of the learning here.
Can I retry the exercise if I want a better score?
Yes — use the 'Try again' button on the results screen to reset your score and go through the questions again. There's no limit on attempts.
Who is this ML Language exercise for?
It's aimed at IT professionals with working English who want to sound more natural and precise around ml language — useful whether you're preparing for real conversations at work or just building confidence with the vocabulary.
Do I need an account to track my progress?
No account is needed. Your progress through the exercise is tracked locally in your browser for the current session, and you can replay the module at any time.
How is this different from reading a blog article?
This exercise is an interactive drill that tests and reinforces specific phrasing through multiple-choice questions with instant feedback, while blog articles explain concepts and vocabulary in prose. The two work well together.
Where can I find more ML Language exercises?
See the ML Language hub for more modules like this one, or browse the full Exercises page for other IT-English topics.
Can I complete this exercise on my phone?
Yes — every exercise on CoderSlingo is fully responsive and works on phones and tablets, so you can practise anywhere.