MLS-C01 Modeling Practice Question

A data scientist is building a text classification model using a bag-of-words approach. The dataset contains 100,000 documents with a vocabulary of 50,000 unique words. The model is overfitting. Which THREE techniques can help reduce overfitting? (Choose THREE.)

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

✓

Apply L1 or L2 regularization

Regularization (L1/L2), reducing n-gram range to unigrams, and feature selection (removing rare words) all reduce model complexity and help prevent overfitting. Option A (increasing max_features) increases complexity and can worsen overfitting. Option E (TF-IDF) is a weighting scheme, not a regularization technique.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
Increase max_features to include more words
Why it's wrong here
Adding more features increases model complexity and overfitting risk.
✓
Apply L1 or L2 regularization
Why this is correct
Regularization penalizes large coefficients, reducing overfitting.
✓
Reduce the n-gram range to unigrams only
Why this is correct
Lower n-gram range reduces feature space and overfitting.
✓
Use feature selection to remove rare words
Why this is correct
Removing rare words reduces noise and overfitting.
✗
Use TF-IDF instead of raw counts
Why it's wrong here
TF-IDF is a feature scaling method, not a regularization technique.

Go deeper

MLS-C01 Modeling Practice Question

Why each option matters

Option-by-option breakdown

Related to this question