easyMultiple Choice
MLA-C01 Practice Question: A machine learning engineer is preparing a…
A machine learning engineer is preparing a dataset for binary classification. The target variable has a severe class imbalance (95% negative, 5% positive). Which technique can help address this imbalance during data preparation?
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
SMOTE (Synthetic Minority Over-sampling Technique)
SMOTE (Synthetic Minority Over-sampling Technique) generates synthetic samples for the minority class, which directly addresses class imbalance by creating more balanced training data. L1 Regularization (Lasso) is a feature selection method, not for imbalance. StandardScaler normalizes features and does not affect class distribution. PCA is a dimensionality reduction technique and does not address imbalance.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
L1 Regularization (Lasso)
Why it's wrong here
L1 regularization penalises absolute coefficient magnitudes to drive feature weights to zero, which addresses overfitting or performs feature selection — it does not alter the 95:5 class distribution. It is tempting because it modifies model training behaviour, and it would be the right choice when irrelevant or redundant features inflate variance rather than when class frequencies need rebalancing.
- ✗
StandardScaler
Why it's wrong here
StandardScaler rescales each feature to zero mean and unit variance; it leaves the 95:5 class ratio untouched, so the classifier still sees the same imbalance. It is tempting because feature scaling genuinely helps distance-based and gradient-descent algorithms converge, and it would be the right preprocessing step when features differ wildly in magnitude.
- ✗
Principal Component Analysis (PCA)
Why it's wrong here
PCA reduces dimensionality by projecting features onto principal components; it does not alter the ratio of positive to negative labels, so the 95/5 split persists. It is tempting because PCA is a standard preprocessing step, and it would be the right choice when many correlated features cause overfitting or slow training.
- ✓
SMOTE (Synthetic Minority Over-sampling Technique)
Why this is correct
SMOTE generates synthetic minority-class examples by interpolating between existing positive instances and their nearest neighbours, rebalancing the 95:5 split. This gives the classifier enough positive signal to learn the minority pattern instead of defaulting to the majority class.
Go deeper
Related to this question
About these practice questions
This MLA-C01 question is part of Courseiva's 665-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This MLA-C01 practice question is part of Courseiva's free Amazon Web Services certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the MLA-C01 exam.