A machine learning engineer is building a spam filter. They have a large dataset of emails, each labeled 'spam' or 'not spam'. They split the data into training and test sets, train a classifier on the training set, and evaluate it on the test set. The test accuracy is high, but when deployed, the filter performs poorly on new emails. Which practice would most likely improve the reliability of the evaluation?

Use k-fold cross-validation on the training data and tune hyperparameters based on validation folds

Increase the size of the test set to 90% of the data

Switch to an unsupervised learning algorithm to avoid label dependency

Train on the test set as well to use all available data

In supervised learning, what is the primary purpose of a labeled testing dataset?

To train the model by adjusting its parameters

To evaluate the model's performance on unseen data

To provide additional data for feature engineering

To label the unlabeled training data

A data scientist is building a model to predict whether a loan applicant will default within the next 12 months. The dataset contains 10,000 labeled examples, with 9,500 non-defaults and 500 defaults. The business prioritizes identifying as many actual defaulters as possible, even if it means flagging some non-defaulters for review. Which algorithm and evaluation metric combination best aligns with this business priority?

Decision tree with precision as the primary metric

Logistic regression with accuracy as the primary metric

Linear regression with F1-score as the primary metric

Support vector machine with recall as the primary metric

Which supervised learning algorithm is specifically designed for classification tasks and works by finding the hyperplane that best separates classes?

Principal Component Analysis (PCA)

GIAC Machine Learning Engineer

Domain 2Objective 3

Supervised Learning GMLE Practice Questions (Page 8)

Part of the Machine Learning Algorithms domain, which makes up ~29% of our current practice bank. GIAC (SANS) does not publish an official question count, but from its 180-minute exam (~70–120 total, ~20–35 in this domain), expect 7–12 from this objective — we provide 40 practice questions to prepare you well beyond it. (estimate)

40questions here

8free pages

6concepts

Questions 36–40

36application · medium
A machine learning engineer is building a spam filter. They have a large dataset of emails, each labeled 'spam' or 'not spam'. They split the data into training and test sets, train a classifier on the training set, and evaluate it on the test set. The test accuracy is high, but when deployed, the filter performs poorly on new emails. Which practice would most likely improve the reliability of the evaluation?
Select an answer first
37foundation · easy
In supervised learning, what is the primary purpose of a labeled testing dataset?
Select an answer first
38foundation · easy
Which technique is commonly used to mitigate overfitting by adding a penalty term to the loss function?
Select an answer first
39application · medium
A data scientist is building a model to predict whether a loan applicant will default within the next 12 months. The dataset contains 10,000 labeled examples, with 9,500 non-defaults and 500 defaults. The business prioritizes identifying as many actual defaulters as possible, even if it means flagging some non-defaulters for review. Which algorithm and evaluation metric combination best aligns with this business priority?
Select an answer first
40foundation · easy
Which supervised learning algorithm is specifically designed for classification tasks and works by finding the hyperplane that best separates classes?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

No more pagesBack to GMLE

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by GIAC (SANS). “GMLE” is a trademark of its owner, used for identification only.

Supervised Learning GMLE Practice Questions (Page 8)

Questions 36–40

In supervised learning, what is the primary purpose of a labeled testing dataset?

Which technique is commonly used to mitigate overfitting by adding a penalty term to the loss function?

Which supervised learning algorithm is specifically designed for classification tasks and works by finding the hyperplane that best separates classes?

Keep studying — Machine Learning Algorithms