A team is working on a dataset where they suspect mislabeled instances. They want to use a statistical outlier analysis to detect them. Which of the following best describes this approach?

Randomly sample instances and have experts review them

Train a model and look at instances where the model's loss is high

Compare the labels to an external authoritative source

Identify instances whose features are significantly different from the majority of instances in the same class

A machine learning engineer is training a model to classify images of animals. They suspect that some images are mislabeled, but they do not know which ones. They decide to use a model-based approach to detect mislabeled data. Which approach best fits this description?

Train a model on the dataset and examine the instances where the model's predictions have low confidence

Manually inspect a random sample of images to look for labeling errors

Calculate the average color histogram of each image and compare it to the class average

Remove all images that are smaller than a certain resolution

Which of the following is a common source of mislabeled data in automated labeling pipelines?

A rule-based system that incorrectly assigns labels to certain edge cases

Manual review of a random sample of labels by a domain expert

Data augmentation that flips images horizontally

Using a pre-trained model to generate labels with high confidence

A logistics company is building a model to predict package delivery delays. The training data includes labels that were automatically generated by a rule-based system that flags packages as 'delayed' if they are not scanned within 24 hours. However, some packages are delayed due to weather but are not flagged because they were scanned at a regional hub. What is the primary source of mislabeled data in this scenario?

Human error during manual data entry

Sensor noise in the scanning devices

Automated labeling errors due to incomplete rules

Missing data for packages that were not scanned

Which of the following is an example of mislabeled data, as opposed to another type of data quality issue?

A dataset contains the same record twice

An image of a cat is labeled as 'dog' in the training set

A feature column has values in different units (e.g., meters and feet)

A customer record has a missing phone number

Certified Tester AI Testing

Domain 4Objective 7

Mislabeled Data in Datasets CT-AI Practice Questions (Page 4)

Part of the Domain 4: ML - Data domain, which makes up ~12% of our current practice bank. ISTQB does not publish an official question count, but from its 60-minute exam (~25–40 total, ~3–5 in this domain), expect 1–1 from this objective — we provide 24 practice questions to prepare you well beyond it. (estimate)

24questions here

5free pages

5concepts

Questions 16–20

16application · medium
A team is working on a dataset where they suspect mislabeled instances. They want to use a statistical outlier analysis to detect them. Which of the following best describes this approach?
Select an answer first
17application · medium
A machine learning engineer is training a model to classify images of animals. They suspect that some images are mislabeled, but they do not know which ones. They decide to use a model-based approach to detect mislabeled data. Which approach best fits this description?
Select an answer first
18foundation · easy
Which of the following is a common source of mislabeled data in automated labeling pipelines?
Select an answer first
19application · medium
A logistics company is building a model to predict package delivery delays. The training data includes labels that were automatically generated by a rule-based system that flags packages as 'delayed' if they are not scanned within 24 hours. However, some packages are delayed due to weather but are not flagged because they were scanned at a regional hub. What is the primary source of mislabeled data in this scenario?
Select an answer first
20foundation · easy
Which of the following is an example of mislabeled data, as opposed to another type of data quality issue?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to CT-AI

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 ISTQB. “CT-AI” is a trademark of its owner, used for identification only.